Datasets:

CShorten

/

ML-ArXiv-Papers

like 45

Long-term stability and generalization of observationally-constrained stochastic data-driven models for geophysical turbulence

Recent years have seen a surge in interest in building deep learning-based fully data-driven models for weather prediction. Such deep learning models if trained on observations can mitigate certain biases in current state-of-the-art weather models, some of which stem from inaccurate representation of subgrid-scale processes. However, these data-driven models, being over-parameterized, require a lot of training data which may not be available from reanalysis (observational data) products. Moreover, an accurate, noise-free, initial condition to start forecasting with a data-driven weather model is not available in realistic scenarios. Finally, deterministic data-driven forecasting models suffer from issues with long-term stability and unphysical climate drift, which makes these data-driven models unsuitable for computing climate statistics. Given these challenges, previous studies have tried to pre-train deep learning-based weather forecasting models on a large amount of imperfect long-term climate model simulations and then re-train them on available observational data. In this paper, we propose a convolutional variational autoencoder-based stochastic data-driven model that is pre-trained on an imperfect climate model simulation from a 2-layer quasi-geostrophic flow and re-trained, using transfer learning, on a small number of noisy observations from a perfect simulation. This re-trained model then performs stochastic forecasting with a noisy initial condition sampled from the perfect simulation. We show that our ensemble-based stochastic data-driven model outperforms a baseline deterministic encoder-decoder-based convolutional model in terms of short-term skills while remaining stable for long-term climate simulations yielding accurate climatology.

Affective Medical Estimation and Decision Making via Visualized Learning and Deep Learning

With the advent of sophisticated machine learning (ML) techniques and the promising results they yield, especially in medical applications, where they have been investigated for different tasks to enhance the decision-making process. Since visualization is such an effective tool for human comprehension, memorization, and judgment, we have presented a first-of-its-kind estimation approach we refer to as Visualized Learning for Machine Learning (VL4ML) that not only can serve to assist physicians and clinicians in making reasoned medical decisions, but it also allows to appreciate the uncertainty visualization, which could raise incertitude in making the appropriate classification or prediction. For the proof of concept, and to demonstrate the generalized nature of this visualized estimation approach, five different case studies are examined for different types of tasks including classification, regression, and longitudinal prediction. A survey analysis with more than 100 individuals is also conducted to assess users' feedback on this visualized estimation method. The experiments and the survey demonstrate the practical merits of the VL4ML that include: (1) appreciating visually clinical/medical estimations; (2) getting closer to the patients' preferences; (3) improving doctor-patient communication, and (4) visualizing the uncertainty introduced through the black box effect of the deployed ML algorithm. All the source codes are shared via a GitHub repository.

A Verification Framework for Certifying Learning-Based Safety-Critical Aviation Systems

We present a safety verification framework for design-time and run-time assurance of learning-based components in aviation systems. Our proposed framework integrates two novel methodologies. From the design-time assurance perspective, we propose offline mixed-fidelity verification tools that incorporate knowledge from different levels of granularity in simulated environments. From the run-time assurance perspective, we propose reachability- and statistics-based online monitoring and safety guards for a learning-based decision-making model to complement the offline verification methods. This framework is designed to be loosely coupled among modules, allowing the individual modules to be developed using independent methodologies and techniques, under varying circumstances and with different tool access. The proposed framework offers feasible solutions for meeting system safety requirements at different stages throughout the system development and deployment cycle, enabling the continuous learning and assessment of the system product.

Towards Optimal VPU Compiler Cost Modeling by using Neural Networks to Infer Hardware Performances

Calculating the most efficient schedule of work in a neural network compiler is a difficult task. There are many parameters to be accounted for that can positively or adversely affect that schedule depending on their configuration - How work is shared between distributed targets, the subdivision of tensors to fit in memory, toggling the enablement of optimizations, etc. Traditionally, neural network compilers determine how to set these values by building a graph of choices and choosing the path with minimal 'cost'. These choices and their corresponding costs are usually determined by an algorithm crafted by engineers with a deep knowledge of the target platform. However, when the amount of options available to a compiler is large, it is very difficult to ensure that these models consistently produce an optimal schedule for all scenarios, whilst still completing compilation in an acceptable timeframe. This paper presents 'VPUNN' - a neural network-based cost model trained on low-level task profiling that consistently outperforms the state-of-the-art cost modeling in Intel's line of VPU processors.

A Song of (Dis)agreement: Evaluating the Evaluation of Explainable Artificial Intelligence in Natural Language Processing

There has been significant debate in the NLP community about whether or not attention weights can be used as an explanation - a mechanism for interpreting how important each input token is for a particular prediction. The validity of "attention as explanation" has so far been evaluated by computing the rank correlation between attention-based explanations and existing feature attribution explanations using LSTM-based models. In our work, we (i) compare the rank correlation between five more recent feature attribution methods and two attention-based methods, on two types of NLP tasks, and (ii) extend this analysis to also include transformer-based models. We find that attention-based explanations do not correlate strongly with any recent feature attribution methods, regardless of the model or task. Furthermore, we find that none of the tested explanations correlate strongly with one another for the transformer-based model, leading us to question the underlying assumption that we should measure the validity of attention-based explanations based on how well they correlate with existing feature attribution explanation methods. After conducting experiments on five datasets using two different models, we argue that the community should stop using rank correlation as an evaluation metric for attention-based explanations. We suggest that researchers and practitioners should instead test various explanation methods and employ a human-in-the-loop process to determine if the explanations align with human intuition for the particular use case at hand.

Should attention be all we need? The epistemic and ethical implications of unification in machine learning

"Attention is all you need" has become a fundamental precept in machine learning research. Originally designed for machine translation, transformers and the attention mechanisms that underpin them now find success across many problem domains. With the apparent domain-agnostic success of transformers, many researchers are excited that similar model architectures can be successfully deployed across diverse applications in vision, language and beyond. We consider the benefits and risks of these waves of unification on both epistemic and ethical fronts. On the epistemic side, we argue that many of the arguments in favor of unification in the natural sciences fail to transfer over to the machine learning case, or transfer over only under assumptions that might not hold. Unification also introduces epistemic risks related to portability, path dependency, methodological diversity, and increased black-boxing. On the ethical side, we discuss risks emerging from epistemic concerns, further marginalizing underrepresented perspectives, the centralization of power, and having fewer models across more domains of application

A Probabilistic Generative Model of Free Categories

Applied category theory has recently developed libraries for computing with morphisms in interesting categories, while machine learning has developed ways of learning programs in interesting languages. Taking the analogy between categories and languages seriously, this paper defines a probabilistic generative model of morphisms in free monoidal categories over domain-specific generating objects and morphisms. The paper shows how acyclic directed wiring diagrams can model specifications for morphisms, which the model can use to generate morphisms. Amortized variational inference in the generative model then enables learning of parameters (by maximum likelihood) and inference of latent variables (by Bayesian inversion). A concrete experiment shows that the free category prior achieves competitive reconstruction performance on the Omniglot dataset.

How Does Frequency Bias Affect the Robustness of Neural Image Classifiers against Common Corruption and Adversarial Perturbations?

Model robustness is vital for the reliable deployment of machine learning models in real-world applications. Recent studies have shown that data augmentation can result in model over-relying on features in the low-frequency domain, sacrificing performance against low-frequency corruptions, highlighting a connection between frequency and robustness. Here, we take one step further to more directly study the frequency bias of a model through the lens of its Jacobians and its implication to model robustness. To achieve this, we propose Jacobian frequency regularization for models' Jacobians to have a larger ratio of low-frequency components. Through experiments on four image datasets, we show that biasing classifiers towards low (high)-frequency components can bring performance gain against high (low)-frequency corruption and adversarial perturbation, albeit with a tradeoff in performance for low (high)-frequency corruption. Our approach elucidates a more direct connection between the frequency bias and robustness of deep learning models.

Selectively Contextual Bandits

Contextual bandits are widely used in industrial personalization systems. These online learning frameworks learn a treatment assignment policy in the presence of treatment effects that vary with the observed contextual features of the users. While personalization creates a rich user experience that reflect individual interests, there are benefits of a shared experience across a community that enable participation in the zeitgeist. Such benefits are emergent through network effects and are not captured in regret metrics typically employed in evaluating bandits. To balance these needs, we propose a new online learning algorithm that preserves benefits of personalization while increasing the commonality in treatments across users. Our approach selectively interpolates between a contextual bandit algorithm and a context-free multi-arm bandit and leverages the contextual information for a treatment decision only if it promises significant gains. Apart from helping users of personalization systems balance their experience between the individualized and shared, simplifying the treatment assignment policy by making it selectively reliant on the context can help improve the rate of learning in some cases. We evaluate our approach in a classification setting using public datasets and show the benefits of the hybrid policy.

Towards a multi-stakeholder value-based assessment framework for algorithmic systems

In an effort to regulate Machine Learning-driven (ML) systems, current auditing processes mostly focus on detecting harmful algorithmic biases. While these strategies have proven to be impactful, some values outlined in documents dealing with ethics in ML-driven systems are still underrepresented in auditing processes. Such unaddressed values mainly deal with contextual factors that cannot be easily quantified. In this paper, we develop a value-based assessment framework that is not limited to bias auditing and that covers prominent ethical principles for algorithmic systems. Our framework presents a circular arrangement of values with two bipolar dimensions that make common motivations and potential tensions explicit. In order to operationalize these high-level principles, values are then broken down into specific criteria and their manifestations. However, some of these value-specific criteria are mutually exclusive and require negotiation. As opposed to some other auditing frameworks that merely rely on ML researchers' and practitioners' input, we argue that it is necessary to include stakeholders that present diverse standpoints to systematically negotiate and consolidate value and criteria tensions. To that end, we map stakeholders with different insight needs, and assign tailored means for communicating value manifestations to them. We, therefore, contribute to current ML auditing practices with an assessment framework that visualizes closeness and tensions between values and we give guidelines on how to operationalize them, while opening up the evaluation and deliberation process to a wide range of stakeholders.

Surreal-GAN:Semi-Supervised Representation Learning via GAN for uncovering heterogeneous disease-related imaging patterns

A plethora of machine learning methods have been applied to imaging data, enabling the construction of clinically relevant imaging signatures of neurological and neuropsychiatric diseases. Oftentimes, such methods don't explicitly model the heterogeneity of disease effects, or approach it via nonlinear models that are not interpretable. Moreover, unsupervised methods may parse heterogeneity that is driven by nuisance confounding factors that affect brain structure or function, rather than heterogeneity relevant to a pathology of interest. On the other hand, semi-supervised clustering methods seek to derive a dichotomous subtype membership, ignoring the truth that disease heterogeneity spatially and temporally extends along a continuum. To address the aforementioned limitations, herein, we propose a novel method, termed Surreal-GAN (Semi-SUpeRvised ReprEsentAtion Learning via GAN). Using cross-sectional imaging data, Surreal-GAN dissects underlying disease-related heterogeneity under the principle of semi-supervised clustering (cluster mappings from normal control to patient), proposes a continuously dimensional representation, and infers the disease severity of patients at individual level along each dimension. The model first learns a transformation function from normal control (CN) domain to the patient (PT) domain with latent variables controlling transformation directions. An inverse mapping function together with regularization on function continuity, pattern orthogonality and monotonicity was also imposed to make sure that the transformation function captures necessarily meaningful imaging patterns with clinical significance. We first validated the model through extensive semi-synthetic experiments, and then demonstrate its potential in capturing biologically plausible imaging patterns in Alzheimer's disease (AD).

Image2Gif: Generating Continuous Realistic Animations with Warping NODEs

Generating smooth animations from a limited number of sequential observations has a number of applications in vision. For example, it can be used to increase number of frames per second, or generating a new trajectory only based on first and last frames, e.g. a motion of face emotions. Despite the discrete observed data (frames), the problem of generating a new trajectory is a continues problem. In addition, to be perceptually realistic, the domain of an image should not alter drastically through the trajectory of changes. In this paper, we propose a new framework, Warping Neural ODE, for generating a smooth animation (video frame interpolation) in a continuous manner, given two ("farther apart") frames, denoting the start and the end of the animation. The key feature of our framework is utilizing the continuous spatial transformation of the image based on the vector field, derived from a system of differential equations. This allows us to achieve the smoothness and the realism of an animation with infinitely small time steps between the frames. We show the application of our work in generating an animation given two frames, in different training settings, including Generative Adversarial Network (GAN) and with $L_2$ loss.

PinnerFormer: Sequence Modeling for User Representation at Pinterest

Sequential models have become increasingly popular in powering personalized recommendation systems over the past several years. These approaches traditionally model a user's actions on a website as a sequence to predict the user's next action. While theoretically simplistic, these models are quite challenging to deploy in production, commonly requiring streaming infrastructure to reflect the latest user activity and potentially managing mutable data for encoding a user's hidden state. Here we introduce PinnerFormer, a user representation trained to predict a user's future long-term engagement using a sequential model of a user's recent actions. Unlike prior approaches, we adapt our modeling to a batch infrastructure via our new dense all-action loss, modeling long-term future actions instead of next action prediction. We show that by doing so, we significantly close the gap between batch user embeddings that are generated once a day and realtime user embeddings generated whenever a user takes an action. We describe our design decisions via extensive offline experimentation and ablations and validate the efficacy of our approach in A/B experiments showing substantial improvements in Pinterest's user retention and engagement when comparing PinnerFormer against our previous user representation. PinnerFormer is deployed in production as of Fall 2021.

Statistical Guarantees for Approximate Stationary Points of Simple Neural Networks

Since statistical guarantees for neural networks are usually restricted to global optima of intricate objective functions, it is not clear whether these theories really explain the performances of actual outputs of neural-network pipelines. The goal of this paper is, therefore, to bring statistical theory closer to practice. We develop statistical guarantees for simple neural networks that coincide up to logarithmic factors with the global optima but apply to stationary points and the points nearby. These results support the common notion that neural networks do not necessarily need to be optimized globally from a mathematical perspective. More generally, despite being limited to simple neural networks for now, our theories make a step forward in describing the practical properties of neural networks in mathematical terms.

Are Quantum Computers Practical Yet? A Case for Feature Selection in Recommender Systems using Tensor Networks

Collaborative filtering models generally perform better than content-based filtering models and do not require careful feature engineering. However, in the cold-start scenario collaborative information may be scarce or even unavailable, whereas the content information may be abundant, but also noisy and expensive to acquire. Thus, selection of particular features that improve cold-start recommendations becomes an important and non-trivial task. In the recent approach by Nembrini et al., the feature selection is driven by the correlational compatibility between collaborative and content-based models. The problem is formulated as a Quadratic Unconstrained Binary Optimization (QUBO) which, due to its NP-hard complexity, is solved using Quantum Annealing on a quantum computer provided by D-Wave. Inspired by the reported results, we contend the idea that current quantum annealers are superior for this problem and instead focus on classical algorithms. In particular, we tackle QUBO via TTOpt, a recently proposed black-box optimizer based on tensor networks and multilinear algebra. We show the computational feasibility of this method for large problems with thousands of features, and empirically demonstrate that the solutions found are comparable to the ones obtained with D-Wave across all examined datasets.

Insights into the origin of halo mass profiles from machine learning

The mass distribution of dark matter haloes is the result of the hierarchical growth of initial density perturbations through mass accretion and mergers. We use an interpretable machine-learning framework to provide physical insights into the origin of the spherically-averaged mass profile of dark matter haloes. We train a gradient-boosted-trees algorithm to predict the final mass profiles of cluster-sized haloes, and measure the importance of the different inputs provided to the algorithm. We find two primary scales in the initial conditions (ICs) that impact the final mass profile: the density at approximately the scale of the haloes' Lagrangian patch $R_L$ ($R\sim 0.7\, R_L$) and that in the large-scale environment ($R\sim 1.7~R_L$). The model also identifies three primary time-scales in the halo assembly history that affect the final profile: (i) the formation time of the virialized, collapsed material inside the halo, (ii) the dynamical time, which captures the dynamically unrelaxed, infalling component of the halo over its first orbit, (iii) a third, most recent time-scale, which captures the impact on the outer profile of recent massive merger events. While the inner profile retains memory of the ICs, this information alone is insufficient to yield accurate predictions for the outer profile. As we add information about the haloes' mass accretion history, we find a significant improvement in the predicted profiles at all radii. Our machine-learning framework provides novel insights into the role of the ICs and the mass assembly history in determining the final mass profile of cluster-sized haloes.

AdaCap: Adaptive Capacity control for Feed-Forward Neural Networks

The capacity of a ML model refers to the range of functions this model can approximate. It impacts both the complexity of the patterns a model can learn but also memorization, the ability of a model to fit arbitrary labels. We propose Adaptive Capacity (AdaCap), a training scheme for Feed-Forward Neural Networks (FFNN). AdaCap optimizes the capacity of FFNN so it can capture the high-level abstract representations underlying the problem at hand without memorizing the training dataset. AdaCap is the combination of two novel ingredients, the Muddling labels for Regularization (MLR) loss and the Tikhonov operator training scheme. The MLR loss leverages randomly generated labels to quantify the propensity of a model to memorize. We prove that the MLR loss is an accurate in-sample estimator for out-of-sample generalization performance and that it can be used to perform Hyper-Parameter Optimization provided a Signal-to-Noise Ratio condition is met. The Tikhonov operator training scheme modulates the capacity of a FFNN in an adaptive, differentiable and data-dependent manner. We assess the effectiveness of AdaCap in a setting where DNN are typically prone to memorization, small tabular datasets, and benchmark its performance against popular machine learning methods.

Introspective Deep Metric Learning

This paper proposes an introspective deep metric learning (IDML) framework for uncertainty-aware comparisons of images. Conventional deep metric learning methods produce confident semantic distances between images regardless of the uncertainty level. However, we argue that a good similarity model should consider the semantic discrepancies with caution to better deal with ambiguous images for more robust training. To achieve this, we propose to represent an image using not only a semantic embedding but also an accompanying uncertainty embedding, which describes the semantic characteristics and ambiguity of an image, respectively. We further propose an introspective similarity metric to make similarity judgments between images considering both their semantic differences and ambiguities. Our framework attains state-of-the-art performance on the widely used CUB-200-2011, Cars196, and Stanford Online Products datasets for image retrieval. We further evaluate our framework for image classification on the ImageNet-1K, CIFAR-10, and CIFAR-100 datasets, which shows that equipping existing data mixing methods with the proposed introspective metric consistently achieves better results (e.g., +0.44 for CutMix on ImageNet-1K). Code is available at: https://github.com/wangck20/IDML.

MixAugment & Mixup: Augmentation Methods for Facial Expression Recognition

Automatic Facial Expression Recognition (FER) has attracted increasing attention in the last 20 years since facial expressions play a central role in human communication. Most FER methodologies utilize Deep Neural Networks (DNNs) that are powerful tools when it comes to data analysis. However, despite their power, these networks are prone to overfitting, as they often tend to memorize the training data. What is more, there are not currently a lot of in-the-wild (i.e. in unconstrained environment) large databases for FER. To alleviate this issue, a number of data augmentation techniques have been proposed. Data augmentation is a way to increase the diversity of available data by applying constrained transformations on the original data. One such technique, which has positively contributed to various classification tasks, is Mixup. According to this, a DNN is trained on convex combinations of pairs of examples and their corresponding labels. In this paper, we examine the effectiveness of Mixup for in-the-wild FER in which data have large variations in head poses, illumination conditions, backgrounds and contexts. We then propose a new data augmentation strategy which is based on Mixup, called MixAugment. According to this, the network is trained concurrently on a combination of virtual examples and real examples; all these examples contribute to the overall loss function. We conduct an extensive experimental study that proves the effectiveness of MixAugment over Mixup and various state-of-the-art methods. We further investigate the combination of dropout with Mixup and MixAugment, as well as the combination of other data augmentation techniques with MixAugment.

BLINK with Elasticsearch for Efficient Entity Linking in Business Conversations

An Entity Linking system aligns the textual mentions of entities in a text to their corresponding entries in a knowledge base. However, deploying a neural entity linking system for efficient real-time inference in production environments is a challenging task. In this work, we present a neural entity linking system that connects the product and organization type entities in business conversations to their corresponding Wikipedia and Wikidata entries. The proposed system leverages Elasticsearch to ensure inference efficiency when deployed in a resource limited cloud machine, and obtains significant improvements in terms of inference speed and memory consumption while retaining high accuracy.

Robustness of Humans and Machines on Object Recognition with Extreme Image Transformations

Recent neural network architectures have claimed to explain data from the human visual cortex. Their demonstrated performance is however still limited by the dependence on exploiting low-level features for solving visual tasks. This strategy limits their performance in case of out-of-distribution/adversarial data. Humans, meanwhile learn abstract concepts and are mostly unaffected by even extreme image distortions. Humans and networks employ strikingly different strategies to solve visual tasks. To probe this, we introduce a novel set of image transforms and evaluate humans and networks on an object recognition task. We found performance for a few common networks quickly decreases while humans are able to recognize objects with a high accuracy.

Accelerated Reinforcement Learning for Temporal Logic Control Objectives

This paper addresses the problem of learning control policies for mobile robots modeled as unknown Markov Decision Processes (MDPs) that are tasked with temporal logic missions, such as sequencing, coverage, or surveillance. The MDP captures uncertainty in the workspace structure and the outcomes of control decisions. The control objective is to synthesize a control policy that maximizes the probability of accomplishing a high-level task, specified as a Linear Temporal Logic (LTL) formula. To address this problem, we propose a novel accelerated model-based reinforcement learning (RL) algorithm for LTL control objectives that is capable of learning control policies significantly faster than related approaches. Its sample-efficiency relies on biasing exploration towards directions that may contribute to task satisfaction. This is accomplished by leveraging an automaton representation of the LTL task as well as a continuously learned MDP model. Finally, we provide extensive comparative experiments that demonstrate the sample efficiency of the proposed method against recent temporal logic RL methods.

Graph Neural Networks for Propositional Model Counting

Graph Neural Networks (GNNs) have been recently leveraged to solve several logical reasoning tasks. Nevertheless, counting problems such as propositional model counting (#SAT) are still mostly approached with traditional solvers. Here we tackle this gap by presenting an architecture based on the GNN framework for belief propagation (BP) of Kuch et al., extended with self-attentive GNN and trained to approximately solve the #SAT problem. We ran a thorough experimental investigation, showing that our model, trained on a small set of random Boolean formulae, is able to scale effectively to much larger problem sizes, with comparable or better performances of state of the art approximate solvers. Moreover, we show that it can be efficiently fine-tuned to provide good generalization results on different formulae distributions, such as those coming from SAT-encoded combinatorial problems.

NaturalSpeech: End-to-End Text to Speech Synthesis with Human-Level Quality

Text to speech (TTS) has made rapid progress in both academia and industry in recent years. Some questions naturally arise that whether a TTS system can achieve human-level quality, how to define/judge that quality and how to achieve it. In this paper, we answer these questions by first defining the human-level quality based on the statistical significance of subjective measure and introducing appropriate guidelines to judge it, and then developing a TTS system called NaturalSpeech that achieves human-level quality on a benchmark dataset. Specifically, we leverage a variational autoencoder (VAE) for end-to-end text to waveform generation, with several key modules to enhance the capacity of the prior from text and reduce the complexity of the posterior from speech, including phoneme pre-training, differentiable duration modeling, bidirectional prior/posterior modeling, and a memory mechanism in VAE. Experiment evaluations on popular LJSpeech dataset show that our proposed NaturalSpeech achieves -0.01 CMOS (comparative mean opinion score) to human recordings at the sentence level, with Wilcoxon signed rank test at p-level p >> 0.05, which demonstrates no statistically significant difference from human recordings for the first time on this dataset.

Model-Contrastive Learning for Backdoor Defense

Due to the popularity of Artificial Intelligence (AI) techniques, we are witnessing an increasing number of backdoor injection attacks that are designed to maliciously threaten Deep Neural Networks (DNNs) causing misclassification. Although there exist various defense methods that can effectively erase backdoors from DNNs, they greatly suffer from both high Attack Success Rate (ASR) and a non-negligible loss in Benign Accuracy (BA). Inspired by the observation that a backdoored DNN tends to form a new cluster in its feature spaces for poisoned data, in this paper we propose a novel two-stage backdoor defense method, named MCLDef, based on Model-Contrastive Learning (MCL). In the first stage, our approach performs trigger inversion based on trigger synthesis, where the resultant trigger can be used to generate poisoned data. In the second stage, under the guidance of MCL and our defined positive and negative pairs, MCLDef can purify the backdoored model by pulling the feature representations of poisoned data towards those of their clean data counterparts. Due to the shrunken cluster of poisoned data, the backdoor formed by end-to-end supervised learning is eliminated. Comprehensive experimental results show that, with only 5% of clean data, MCLDef significantly outperforms state-of-the-art defense methods by up to 95.79% reduction in ASR, while in most cases the BA degradation can be controlled within less than 2%. Our code is available at https://github.com/WeCanShow/MCL.

EigenNoise: A Contrastive Prior to Warm-Start Representations

In this work, we present a naive initialization scheme for word vectors based on a dense, independent co-occurrence model and provide preliminary results that suggest it is competitive and warrants further investigation. Specifically, we demonstrate through information-theoretic minimum description length (MDL) probing that our model, EigenNoise, can approach the performance of empirically trained GloVe despite the lack of any pre-training data (in the case of EigenNoise). We present these preliminary results with interest to set the stage for further investigations into how this competitive initialization works without pre-training data, as well as to invite the exploration of more intelligent initialization schemes informed by the theory of harmonic linguistic structure. Our application of this theory likewise contributes a novel (and effective) interpretation of recent discoveries which have elucidated the underlying distributional information that linguistic representations capture from data and contrast distributions.

Beyond a Pre-Trained Object Detector: Cross-Modal Textual and Visual Context for Image Captioning

Significant progress has been made on visual captioning, largely relying on pre-trained features and later fixed object detectors that serve as rich inputs to auto-regressive models. A key limitation of such methods, however, is that the output of the model is conditioned only on the object detector's outputs. The assumption that such outputs can represent all necessary information is unrealistic, especially when the detector is transferred across datasets. In this work, we reason about the graphical model induced by this assumption, and propose to add an auxiliary input to represent missing information such as object relationships. We specifically propose to mine attributes and relationships from the Visual Genome dataset and condition the captioning model on them. Crucially, we propose (and show to be important) the use of a multi-modal pre-trained model (CLIP) to retrieve such contextual descriptions. Further, object detector models are frozen and do not have sufficient richness to allow the captioning model to properly ground them. As a result, we propose to condition both the detector and description outputs on the image, and show qualitatively and quantitatively that this can improve grounding. We validate our method on image captioning, perform thorough analyses of each component and importance of the pre-trained multi-modal model, and demonstrate significant improvements over the current state of the art, specifically +7.5% in CIDEr and +1.3% in BLEU-4 metrics.

Classification and mapping of low-statured 'shrubland' cover types in post-agricultural landscapes of the US Northeast

Context: Novel plant communities reshape landscapes and pose challenges for land cover classification and mapping that can constrain research and stewardship efforts. In the US Northeast, emergence of low-statured woody vegetation, or 'shrublands', instead of secondary forests in post-agricultural landscapes is well-documented by field studies, but poorly understood from a landscape perspective, which limits the ability to systematically study and manage these lands. Objectives: To address gaps in classification/mapping of low-statured cover types where they have been historically rare, we developed models to predict 'shrubland' distributions at 30m resolution across New York State (NYS), using machine learning and model ensembling techniques to integrate remote sensing of structural (airborne LIDAR) and optical (satellite imagery) properties of vegetation cover. We first classified a 1m canopy height model (CHM), derived from a "patchwork" of available LIDAR coverages, to define shrubland presence/absence. Next, these non-contiguous maps were used to train a model ensemble based on temporally-segmented imagery to predict 'shrubland' probability for the entire study landscape (NYS). Results: Approximately 2.5% of the CHM coverage area was classified as shrubland. Models using Landsat predictors trained on the classified CHM were effective at identifying shrubland (test set AUC=0.893, real-world AUC=0.904), in discriminating between shrub/young forest and other cover classes, and produced qualitatively sensible maps, even when extending beyond the original training data. Conclusions: After ground-truthing, we expect these shrubland maps and models will have many research and stewardship applications including wildlife conservation, invasive species mitigation and natural climate solutions.

Towards Feature Selection for Ranking and Classification Exploiting Quantum Annealers

Feature selection is a common step in many ranking, classification, or prediction tasks and serves many purposes. By removing redundant or noisy features, the accuracy of ranking or classification can be improved and the computational cost of the subsequent learning steps can be reduced. However, feature selection can be itself a computationally expensive process. While for decades confined to theoretical algorithmic papers, quantum computing is now becoming a viable tool to tackle realistic problems, in particular special-purpose solvers based on the Quantum Annealing paradigm. This paper aims to explore the feasibility of using currently available quantum computing architectures to solve some quadratic feature selection algorithms for both ranking and classification. The experimental analysis includes 15 state-of-the-art datasets. The effectiveness obtained with quantum computing hardware is comparable to that of classical solvers, indicating that quantum computers are now reliable enough to tackle interesting problems. In terms of scalability, current generation quantum computers are able to provide a limited speedup over certain classical algorithms and hybrid quantum-classical strategies show lower computational cost for problems of more than a thousand features.

Transfer Learning Based Efficient Traffic Prediction with Limited Training Data

Efficient prediction of internet traffic is an essential part of Self Organizing Network (SON) for ensuring proactive management. There are many existing solutions for internet traffic prediction with higher accuracy using deep learning. But designing individual predictive models for each service provider in the network is challenging due to data heterogeneity, scarcity, and abnormality. Moreover, the performance of the deep sequence model in network traffic prediction with limited training data has not been studied extensively in the current works. In this paper, we investigated and evaluated the performance of the deep transfer learning technique in traffic prediction with inadequate historical data leveraging the knowledge of our pre-trained model. First, we used a comparatively larger real-world traffic dataset for source domain prediction based on five different deep sequence models: Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), LSTM Encoder-Decoder (LSTM_En_De), LSTM_En_De with Attention layer (LSTM_En_De_Atn), and Gated Recurrent Unit (GRU). Then, two best-performing models, LSTM_En_De and LSTM_En_De_Atn, from the source domain with an accuracy of 96.06% and 96.05% are considered for the target domain prediction. Finally, four smaller traffic datasets collected for four particular sources and destination pairs are used in the target domain to compare the performance of the standard learning and transfer learning in terms of accuracy and execution time. According to our experimental result, transfer learning helps to reduce the execution time for most cases, while the model's accuracy is improved in transfer learning with a larger training session.

Fatigue Prediction in Outdoor Running Conditions using Audio Data

Although running is a common leisure activity and a core training regiment for several athletes, between $29\%$ and $79\%$ of runners sustain an overuse injury each year. These injuries are linked to excessive fatigue, which alters how someone runs. In this work, we explore the feasibility of modelling the Borg received perception of exertion (RPE) scale (range: $[6-20]$), a well-validated subjective measure of fatigue, using audio data captured in realistic outdoor environments via smartphones attached to the runners' arms. Using convolutional neural networks (CNNs) on log-Mel spectrograms, we obtain a mean absolute error of $2.35$ in subject-dependent experiments, demonstrating that audio can be effectively used to model fatigue, while being more easily and non-invasively acquired than by signals from other sensors.

Wavelet-Based Hybrid Machine Learning Model for Out-of-distribution Internet Traffic Prediction

Efficient prediction of internet traffic is essential for ensuring proactive management of computer networks. Nowadays, machine learning approaches show promising performance in modeling real-world complex traffic. However, most existing works assumed that model training and evaluation data came from identical distribution. But in practice, there is a high probability that the model will deal with data from a slightly or entirely unknown distribution in the deployment phase. This paper investigated and evaluated machine learning performances using eXtreme Gradient Boosting, Light Gradient Boosting Machine, Stochastic Gradient Descent, Gradient Boosting Regressor, CatBoost Regressor, and their stacked ensemble model using data from both identical and out-of distribution. Also, we proposed a hybrid machine learning model integrating wavelet decomposition for improving out-of-distribution prediction as standalone models were unable to generalize very well. Our experimental results show the best performance of the standalone ensemble model with an accuracy of 96.4%, while the hybrid ensemble model improved it by 1% for in-distribution data. But its performance dropped significantly when tested with three different datasets having a distribution shift than the training set. However, our proposed hybrid model considerably reduces the performance gap between identical and out-of-distribution evaluation compared with the standalone model, indicating the decomposition technique's effectiveness in the case of out-of-distribution generalization.

Protecting Data from all Parties: Combining FHE and DP in Federated Learning

This paper tackles the problem of ensuring training data privacy in a federated learning context. Relying on Homomorphic Encryption (HE) and Differential Privacy (DP), we propose a framework addressing threats on the privacy of the training data. Notably, the proposed framework ensures the privacy of the training data from all actors of the learning process, namely the data owners and the aggregating server. More precisely, while HE blinds a semi-honest server during the learning protocol, DP protects the data from semi-honest clients participating in the training process as well as end-users with black-box or white-box access to the trained model. In order to achieve this, we provide new theoretical and practical results to allow these techniques to be rigorously combined. In particular, by means of a novel stochastic quantisation operator, we prove DP guarantees in a context where the noise is quantised and bounded due to the use of HE. The paper is concluded by experiments which show the practicality of the entire framework in terms of both model quality (impacted by DP) and computational overhead (impacted by HE).

Insights on Modelling Physiological, Appraisal, and Affective Indicators of Stress using Audio Features

Stress is a major threat to well-being that manifests in a variety of physiological and mental symptoms. Utilising speech samples collected while the subject is undergoing an induced stress episode has recently shown promising results for the automatic characterisation of individual stress responses. In this work, we introduce new findings that shed light onto whether speech signals are suited to model physiological biomarkers, as obtained via cortisol measurements, or self-assessed appraisal and affect measurements. Our results show that different indicators impact acoustic features in a diverse way, but that their complimentary information can nevertheless be effectively harnessed by a multi-tasking architecture to improve prediction performance for all of them.

HierAttn: Effectively Learn Representations from Stage Attention and Branch Attention for Skin Lesions Diagnosis

Accurate and unbiased examinations of skin lesions are critical for the early diagnosis and treatment of skin conditions and disorders. Visual features of skin lesions vary significantly because the images are collected from patients with different lesion colours and morphologies by using dissimilar imaging equipment. Recent studies have reported ensembled convolutional neural networks (CNNs) to classify the images for early diagnosis of skin disorders. However, the practical use of these ensembled CNNs is limited because they are heavyweight and inadequate for using contextual information. Although lightweight networks (e.g., MobileNetV3 and EfficientNet) were developed to achieve parameters reduction for implementing deep neural networks on mobile devices, insufficient depth of feature representation restricts the performance. To address the existing limitations, we introduce a new lite and effective neural network, namely HierAttn. The HierAttn applies a novel strategy to learn the local and global features by using multi-stage and multi-branch attention mechanisms. The efficacy of HierAttn was evaluated by using the dermoscopy images dataset ISIC2019 and smartphone photos dataset PAD-UFES-20 (PAD20). The experimental results show that HierAttn achieves the best accuracy and AUC among the state-of-the-art lightweight networks. The code is available at https://github.com/anthonyweidai/HierAttn.

Evaluating the Fairness Impact of Differentially Private Synthetic Data

Differentially private (DP) synthetic data is a promising approach to maximizing the utility of data containing sensitive information. Due to the suppression of underrepresented classes that is often required to achieve privacy, however, it may be in conflict with fairness. We evaluate four DP synthesizers and present empirical results indicating that three of these models frequently degrade fairness outcomes on downstream binary classification tasks. We draw a connection between fairness and the proportion of minority groups present in the generated synthetic data, and find that training synthesizers on data that are pre-processed via a multi-label undersampling method can promote more fair outcomes without degrading accuracy.

TGANet: Text-guided attention for improved polyp segmentation

Colonoscopy is a gold standard procedure but is highly operator-dependent. Automated polyp segmentation, a precancerous precursor, can minimize missed rates and timely treatment of colon cancer at an early stage. Even though there are deep learning methods developed for this task, variability in polyp size can impact model training, thereby limiting it to the size attribute of the majority of samples in the training dataset that may provide sub-optimal results to differently sized polyps. In this work, we exploit size-related and polyp number-related features in the form of text attention during training. We introduce an auxiliary classification task to weight the text-based embedding that allows network to learn additional feature representations that can distinctly adapt to differently sized polyps and can adapt to cases with multiple polyps. Our experimental results demonstrate that these added text embeddings improve the overall performance of the model compared to state-of-the-art segmentation methods. We explore four different datasets and provide insights for size-specific improvements. Our proposed text-guided attention network (TGANet) can generalize well to variable-sized polyps in different datasets.

Multi-segment preserving sampling for deep manifold sampler

Deep generative modeling for biological sequences presents a unique challenge in reconciling the bias-variance trade-off between explicit biological insight and model flexibility. The deep manifold sampler was recently proposed as a means to iteratively sample variable-length protein sequences by exploiting the gradients from a function predictor. We introduce an alternative approach to this guided sampling procedure, multi-segment preserving sampling, that enables the direct inclusion of domain-specific knowledge by designating preserved and non-preserved segments along the input sequence, thereby restricting variation to only select regions. We present its effectiveness in the context of antibody design by training two models: a deep manifold sampler and a GPT-2 language model on nearly six million heavy chain sequences annotated with the IGHV1-18 gene. During sampling, we restrict variation to only the complementarity-determining region 3 (CDR3) of the input. We obtain log probability scores from a GPT-2 model for each sampled CDR3 and demonstrate that multi-segment preserving sampling generates reasonable designs while maintaining the desired, preserved regions.

Research on the correlation between text emotion mining and stock market based on deep learning

This paper discusses how to crawl the data of financial forums such as stock bar, and conduct emotional analysis combined with the in-depth learning model. This paper will use the Bert model to train the financial corpus and predict the Shenzhen stock index. Through the comparative study of the maximal information coefficient (MIC), it is found that the emotional characteristics obtained by applying the BERT model to the financial corpus can be reflected in the fluctuation of the stock market, which is conducive to effectively improve the prediction accuracy. At the same time, this paper combines in-depth learning with financial texts to further explore the impact mechanism of investor sentiment on the stock market through in-depth learning, which will help the national regulatory authorities and policy departments to formulate more reasonable policies and guidelines for maintaining the stability of the stock market.

An Effective Scheme for Maize Disease Recognition based on Deep Networks

In the last decades, the area under cultivation of maize products has increased because of its essential role in the food cycle for humans, livestock, and poultry. Moreover, the diseases of plants impact food safety and can significantly reduce both the quality and quantity of agricultural products. There are many challenges to accurate and timely diagnosis of the disease. This research presents a novel scheme based on a deep neural network to overcome the mentioned challenges. Due to the limited number of data, the transfer learning technique is employed with the help of two well-known architectures. In this way, a new effective model is adopted by a combination of pre-trained MobileNetV2 and Inception Networks due to their effective performance on object detection problems. The convolution layers of MoblieNetV2 and Inception modules are parallelly arranged as earlier layers to extract crucial features. In addition, the imbalance problem of classes has been solved by an augmentation strategy. The proposed scheme has a superior performance compared to other state-of-the-art models published in recent years. The accuracy of the model reaches 97%, approximately. In summary, experimental results prove the method's validity and significant performance in diagnosing disease in plant leaves.

A Dataset and BERT-based Models for Targeted Sentiment Analysis on Turkish Texts

Targeted Sentiment Analysis aims to extract sentiment towards a particular target from a given text. It is a field that is attracting attention due to the increasing accessibility of the Internet, which leads people to generate an enormous amount of data. Sentiment analysis, which in general requires annotated data for training, is a well-researched area for widely studied languages such as English. For low-resource languages such as Turkish, there is a lack of such annotated data. We present an annotated Turkish dataset suitable for targeted sentiment analysis. We also propose BERT-based models with different architectures to accomplish the task of targeted sentiment analysis. The results demonstrate that the proposed models outperform the traditional sentiment analysis models for the targeted sentiment analysis task.

Attracting and Dispersing: A Simple Approach for Source-free Domain Adaptation

We propose a simple but effective source-free domain adaptation (SFDA) method. Treating SFDA as an unsupervised clustering problem and following the intuition that local neighbors in feature space should have more similar predictions than other features, we propose to optimize an objective of prediction consistency. This objective encourages local neighborhood features in feature space to have similar predictions while features farther away in feature space have dissimilar predictions, leading to efficient feature clustering and cluster assignment simultaneously. For efficient training, we seek to optimize an upper-bound of the objective resulting in two simple terms. Furthermore, we relate popular existing methods in domain adaptation, source-free domain adaptation and contrastive learning via the perspective of discriminability and diversity. The experimental results prove the superiority of our method, and our method can be adopted as a simple but strong baseline for future research in SFDA. Our method can be also adapted to source-free open-set and partial-set DA which further shows the generalization ability of our method.

EF-BV: A Unified Theory of Error Feedback and Variance Reduction Mechanisms for Biased and Unbiased Compression in Distributed Optimization

In distributed or federated optimization and learning, communication between the different computing units is often the bottleneck, and gradient compression is a widely used technique for reducing the number of bits sent within each communication round of iterative methods. There are two classes of compression operators and separate algorithms making use of them. In the case of unbiased random compressors with bounded variance (e.g., rand-k), the DIANA algorithm of Mishchenko et al. [2019], which implements a variance reduction technique for handling the variance introduced by compression, is the current state of the art. In the case of biased and contractive compressors (e.g., top-k), the EF21 algorithm of Richt\'arik et al. [2021], which implements an error-feedback mechanism for handling the error introduced by compression, is the current state of the art. These two classes of compression schemes and algorithms are distinct, with different analyses and proof techniques. In this paper, we unify them into a single framework and propose a new algorithm, recovering DIANA and EF21 as particular cases. We prove linear convergence under certain conditions. Our general approach works with a new, larger class of compressors, which includes unbiased and biased compressors as particular cases, and has two parameters, the bias and the variance. These gives a finer control and allows us to inherit the best of the two worlds: biased compressors, whose good performance in practice is recognized, can be used. And independent randomness at the compressors allows to mitigate the effects of compression, with the convergence rate improving when the number of parallel workers is large. This is the first time that an algorithm with all these features is proposed. Our approach takes a step towards better understanding of two so-far distinct worlds of communication-efficient distributed learning.

Residue-based Label Protection Mechanisms in Vertical Logistic Regression

Federated learning (FL) enables distributed participants to collaboratively learn a global model without revealing their private data to each other. Recently, vertical FL, where the participants hold the same set of samples but with different features, has received increased attention. This paper first presents one label inference attack method to investigate the potential privacy leakages of the vertical logistic regression model. Specifically, we discover that the attacker can utilize the residue variables, which are calculated by solving the system of linear equations constructed by local dataset and the received decrypted gradients, to infer the privately owned labels. To deal with this, we then propose three protection mechanisms, e.g., additive noise mechanism, multiplicative noise mechanism, and hybrid mechanism which leverages local differential privacy and homomorphic encryption techniques, to prevent the attack and improve the robustness of the vertical logistic regression. model. Experimental results show that both the additive noise mechanism and the multiplicative noise mechanism can achieve efficient label protection with only a slight drop in model testing accuracy, furthermore, the hybrid mechanism can achieve label protection without any testing accuracy degradation, which demonstrates the effectiveness and efficiency of our protection techniques

Auto-SDE: Learning effective reduced dynamics from data-driven stochastic dynamical systems

Multiscale stochastic dynamical systems have been widely adopted to scientific and engineering problems due to their capability of depicting complex phenomena in many real world applications. This work is devoted to investigating the effective reduced dynamics for a slow-fast stochastic dynamical system. Given observation data on a short-term period satisfying some unknown slow-fast stochastic system, we propose a novel algorithm including a neural network called Auto-SDE to learn invariant slow manifold. Our approach captures the evolutionary nature of a series of time-dependent autoencoder neural networks with the loss constructed from a discretized stochastic differential equation. Our algorithm is also proved to be accurate, stable and effective through numerical experiments under various evaluation metrics.

Verifying Integrity of Deep Ensemble Models by Lossless Black-box Watermarking with Sensitive Samples

With the widespread use of deep neural networks (DNNs) in many areas, more and more studies focus on protecting DNN models from intellectual property (IP) infringement. Many existing methods apply digital watermarking to protect the DNN models. The majority of them either embed a watermark directly into the internal network structure/parameters or insert a zero-bit watermark by fine-tuning a model to be protected with a set of so-called trigger samples. Though these methods work very well, they were designed for individual DNN models, which cannot be directly applied to deep ensemble models (DEMs) that combine multiple DNN models to make the final decision. It motivates us to propose a novel black-box watermarking method in this paper for DEMs, which can be used for verifying the integrity of DEMs. In the proposed method, a certain number of sensitive samples are carefully selected through mimicking real-world DEM attacks and analyzing the prediction results of the sub-models of the non-attacked DEM and the attacked DEM on the carefully crafted dataset. By analyzing the prediction results of the target DEM on these carefully crafted sensitive samples, we are able to verify the integrity of the target DEM. Different from many previous methods, the proposed method does not modify the original DEM to be protected, which indicates that the proposed method is lossless. Experimental results have shown that the DEM integrity can be reliably verified even if only one sub-model was attacked, which has good potential in practice.

The Roles and Modes of Human Interactions with Automated Machine Learning Systems

As automated machine learning (AutoML) systems continue to progress in both sophistication and performance, it becomes important to understand the `how' and `why' of human-computer interaction (HCI) within these frameworks, both current and expected. Such a discussion is necessary for optimal system design, leveraging advanced data-processing capabilities to support decision-making involving humans, but it is also key to identifying the opportunities and risks presented by ever-increasing levels of machine autonomy. Within this context, we focus on the following questions: (i) How does HCI currently look like for state-of-the-art AutoML algorithms, especially during the stages of development, deployment, and maintenance? (ii) Do the expectations of HCI within AutoML frameworks vary for different types of users and stakeholders? (iii) How can HCI be managed so that AutoML solutions acquire human trust and broad acceptance? (iv) As AutoML systems become more autonomous and capable of learning from complex open-ended environments, will the fundamental nature of HCI evolve? To consider these questions, we project existing literature in HCI into the space of AutoML; this connection has, to date, largely been unexplored. In so doing, we review topics including user-interface design, human-bias mitigation, and trust in artificial intelligence (AI). Additionally, to rigorously gauge the future of HCI, we contemplate how AutoML may manifest in effectively open-ended environments. This discussion necessarily reviews projected developmental pathways for AutoML, such as the incorporation of reasoning, although the focus remains on how and why HCI may occur in such a framework rather than on any implementational details. Ultimately, this review serves to identify key research directions aimed at better facilitating the roles and modes of human interactions with both current and future AutoML systems.

Btech thesis report on adversarial attack detection and purification of adverserially attacked images

This is Btech thesis report on detection and purification of adverserially attacked images. A deep learning model is trained on certain training examples for various tasks such as classification, regression etc. By training, weights are adjusted such that the model performs the task well not only on training examples judged by a certain metric but has an excellent ability to generalize on other unseen examples as well which are typically called the test data. Despite the huge success of machine learning models on a wide range of tasks, security has received a lot less attention along the years. Robustness along various potential cyber attacks also should be a metric for the accuracy of the machine learning models. These cyber attacks can potentially lead to a variety of negative impacts in the real world sensitive applications for which machine learning is used such as medical and transportation systems. Hence, it is a necessity to secure the system from such attacks. Int this report, I focus on a class of these cyber attacks called the adversarial attacks in which the original input sample is modified by small perturbations such that they still look visually the same to human beings but the machine learning models are fooled by such inputs. In this report I discuss 2 novel ways to counter the adversarial attack using AutoEncoders, 1) by detecting the presence of adversaries and 2) purifying these adversaries to make target classification models robust against such attacks.

Federated Multi-Armed Bandits Under Byzantine Attacks

Multi-armed bandits (MAB) is a simple reinforcement learning model where the learner controls the trade-off between exploration versus exploitation to maximize its cumulative reward. Federated multi-armed bandits (FMAB) is a recently emerging framework where a cohort of learners with heterogeneous local models play a MAB game and communicate their aggregated feedback to a parameter server to learn the global feedback model. Federated learning models are vulnerable to adversarial attacks such as model-update attacks or data poisoning. In this work, we study an FMAB problem in the presence of Byzantine clients who can send false model updates that pose a threat to the learning process. We borrow tools from robust statistics and propose a median-of-means-based estimator: Fed-MoM-UCB, to cope with the Byzantine clients. We show that if the Byzantine clients constitute at most half the cohort, it is possible to incur a cumulative regret on the order of ${\cal O} (\log T)$ with respect to an unavoidable error margin, including the communication cost between the clients and the parameter server. We analyze the interplay between the algorithm parameters, unavoidable error margin, regret, communication cost, and the arms' suboptimality gaps. We demonstrate Fed-MoM-UCB's effectiveness against the baselines in the presence of Byzantine attacks via experiments.

Quantum neural network autoencoder and classifier applied to an industrial case study

Quantum computing technologies are in the process of moving from academic research to real industrial applications, with the first hints of quantum advantage demonstrated in recent months. In these early practical uses of quantum computers it is relevant to develop algorithms that are useful for actual industrial processes. In this work we propose a quantum pipeline, comprising a quantum autoencoder followed by a quantum classifier, which are used to first compress and then label classical data coming from a separator, i.e., a machine used in one of Eni's Oil Treatment Plants. This work represents one of the first attempts to integrate quantum computing procedures in a real-case scenario of an industrial pipeline, in particular using actual data coming from physical machines, rather than pedagogical data from benchmark datasets.

Predicting tacrolimus exposure in kidney transplanted patients using machine learning

Tacrolimus is one of the cornerstone immunosuppressive drugs in most transplantation centers worldwide following solid organ transplantation. Therapeutic drug monitoring of tacrolimus is necessary in order to avoid rejection of the transplanted organ or severe side effects. However, finding the right dose for a given patient is challenging, even for experienced clinicians. Consequently, a tool that can accurately estimate the drug exposure for individual dose adaptions would be of high clinical value. In this work, we propose a new technique using machine learning to estimate the tacrolimus exposure in kidney transplant recipients. Our models achieve predictive errors that are at the same level as an established population pharmacokinetic model, but are faster to develop and require less knowledge about the pharmacokinetic properties of the drug.

Localized Adversarial Domain Generalization

Deep learning methods can struggle to handle domain shifts not seen in training data, which can cause them to not generalize well to unseen domains. This has led to research attention on domain generalization (DG), which aims to the model's generalization ability to out-of-distribution. Adversarial domain generalization is a popular approach to DG, but conventional approaches (1) struggle to sufficiently align features so that local neighborhoods are mixed across domains; and (2) can suffer from feature space over collapse which can threaten generalization performance. To address these limitations, we propose localized adversarial domain generalization with space compactness maintenance~(LADG) which constitutes two major contributions. First, we propose an adversarial localized classifier as the domain discriminator, along with a principled primary branch. This constructs a min-max game whereby the aim of the featurizer is to produce locally mixed domains. Second, we propose to use a coding-rate loss to alleviate feature space over collapse. We conduct comprehensive experiments on the Wilds DG benchmark to validate our approach, where LADG outperforms leading competitors on most datasets.

On Generalisability of Machine Learning-based Network Intrusion Detection Systems

Many of the proposed machine learning (ML) based network intrusion detection systems (NIDSs) achieve near perfect detection performance when evaluated on synthetic benchmark datasets. Though, there is no record of if and how these results generalise to other network scenarios, in particular to real-world networks. In this paper, we investigate the generalisability property of ML-based NIDSs by extensively evaluating seven supervised and unsupervised learning models on four recently published benchmark NIDS datasets. Our investigation indicates that none of the considered models is able to generalise over all studied datasets. Interestingly, our results also indicate that the generalisability has a high degree of asymmetry, i.e., swapping the source and target domains can significantly change the classification performance. Our investigation also indicates that overall, unsupervised learning methods generalise better than supervised learning models in our considered scenarios. Using SHAP values to explain these results indicates that the lack of generalisability is mainly due to the presence of strong correspondence between the values of one or more features and Attack/Benign classes in one dataset-model combination and its absence in other datasets that have different feature distributions.

SmoothNets: Optimizing CNN architecture design for differentially private deep learning

The arguably most widely employed algorithm to train deep neural networks with Differential Privacy is DPSGD, which requires clipping and noising of per-sample gradients. This introduces a reduction in model utility compared to non-private training. Empirically, it can be observed that this accuracy degradation is strongly dependent on the model architecture. We investigated this phenomenon and, by combining components which exhibit good individual performance, distilled a new model architecture termed SmoothNet, which is characterised by increased robustness to the challenges of DP-SGD training. Experimentally, we benchmark SmoothNet against standard architectures on two benchmark datasets and observe that our architecture outperforms others, reaching an accuracy of 73.5\% on CIFAR-10 at $\varepsilon=7.0$ and 69.2\% at $\varepsilon=7.0$ on ImageNette, a state-of-the-art result compared to prior architectural modifications for DP.

PS-Net: Deep Partially Separable Modelling for Dynamic Magnetic Resonance Imaging

Deep learning methods driven by the low-rank regularization have achieved attractive performance in dynamic magnetic resonance (MR) imaging. However, most of these methods represent low-rank prior by hand-crafted nuclear norm, which cannot accurately approximate the low-rank prior over the entire dataset through a fixed regularization parameter. In this paper, we propose a learned low-rank method for dynamic MR imaging. In particular, we unrolled the semi-quadratic splitting method (HQS) algorithm for the partially separable (PS) model to a network, in which the low-rank is adaptively characterized by a learnable null-space transform. Experiments on the cardiac cine dataset show that the proposed model outperforms the state-of-the-art compressed sensing (CS) methods and existing deep learning methods both quantitatively and qualitatively.

Augmentations: An Insight into their Effectiveness on Convolution Neural Networks

Augmentations are the key factor in determining the performance of any neural network as they provide a model with a critical edge in boosting its performance. Their ability to boost a model's robustness depends on two factors, viz-a-viz, the model architecture, and the type of augmentations. Augmentations are very specific to a dataset, and it is not imperative that all kinds of augmentation would necessarily produce a positive effect on a model's performance. Hence there is a need to identify augmentations that perform consistently well across a variety of datasets and also remain invariant to the type of architecture, convolutions, and the number of parameters used. Hence there is a need to identify augmentations that perform consistently well across a variety of datasets and also remain invariant to the type of architecture, convolutions, and the number of parameters used. This paper evaluates the effect of parameters using 3x3 and depth-wise separable convolutions on different augmentation techniques on MNIST, FMNIST, and CIFAR10 datasets. Statistical Evidence shows that techniques such as Cutouts and Random horizontal flip were consistent on both parametrically low and high architectures. Depth-wise separable convolutions outperformed 3x3 convolutions at higher parameters due to their ability to create deeper networks. Augmentations resulted in bridging the accuracy gap between the 3x3 and depth-wise separable convolutions, thus establishing their role in model generalization. At higher number augmentations did not produce a significant change in performance. The synergistic effect of multiple augmentations at higher parameters, with antagonistic effect at lower parameters, was also evaluated. The work proves that a delicate balance between architectural supremacy and augmentations needs to be achieved to enhance a model's performance in any given deep learning task.

Exploiting Digital Surface Models for Inferring Super-Resolution for Remotely Sensed Images

Despite the plethora of successful Super-Resolution Reconstruction (SRR) models applied to natural images, their application to remote sensing imagery tends to produce poor results. Remote sensing imagery is often more complicated than natural images and has its peculiarities such as being of lower resolution, it contains noise, and often depicting large textured surfaces. As a result, applying non-specialized SRR models on remote sensing imagery results in artifacts and poor reconstructions. To address these problems, this paper proposes an architecture inspired by previous research work, introducing a novel approach for forcing an SRR model to output realistic remote sensing images: instead of relying on feature-space similarities as a perceptual loss, the model considers pixel-level information inferred from the normalized Digital Surface Model (nDSM) of the image. This strategy allows the application of better-informed updates during the training of the model which sources from a task (elevation map inference) that is closely related to remote sensing. Nonetheless, the nDSM auxiliary information is not required during production and thus the model infers a super-resolution image without any additional data besides its low-resolution pairs. We assess our model on two remotely sensed datasets of different spatial resolutions that also contain the DSM pairs of the images: the DFC2018 dataset and the dataset containing the national Lidar fly-by of Luxembourg. Based on visual inspection, the inferred super-resolution images exhibit particularly superior quality. In particular, the results for the high-resolution DFC2018 dataset are realistic and almost indistinguishable from the ground truth images.

Unsupervised Learning of Rydberg Atom Array Phase Diagram with Siamese Neural Networks

We introduce an unsupervised machine learning method based on Siamese Neural Networks (SNN) to detect phase boundaries. This method is applied to Monte-Carlo simulations of Ising-type systems and Rydberg atom arrays. In both cases the SNN reveals phase boundaries consistent with prior research. The combination of leveraging the power of feed-forward neural networks, unsupervised learning and the ability to learn about multiple phases without knowing about their existence provides a powerful method to explore new and unknown phases of matter.

Masked Co-attentional Transformer reconstructs 100x ultra-fast/low-dose whole-body PET from longitudinal images and anatomically guided MRI

Despite its tremendous value for the diagnosis, treatment monitoring and surveillance of children with cancer, whole body staging with positron emission tomography (PET) is time consuming and associated with considerable radiation exposure. 100x (1% of the standard clinical dosage) ultra-low-dose/ultra-fast whole-body PET reconstruction has the potential for cancer imaging with unprecedented speed and improved safety, but it cannot be achieved by the naive use of machine learning techniques. In this study, we utilize the global similarity between baseline and follow-up PET and magnetic resonance (MR) images to develop Masked-LMCTrans, a longitudinal multi-modality co-attentional CNN-Transformer that provides interaction and joint reasoning between serial PET/MRs of the same patient. We mask the tumor area in the referenced baseline PET and reconstruct the follow-up PET scans. In this manner, Masked-LMCTrans reconstructs 100x almost-zero radio-exposure whole-body PET that was not possible before. The technique also opens a new pathway for longitudinal radiology imaging reconstruction, a significantly under-explored area to date. Our model was trained and tested with Stanford PET/MRI scans of pediatric lymphoma patients and evaluated externally on PET/MRI images from T\"ubingen University. The high image quality of the reconstructed 100x whole-body PET images resulting from the application of Masked-LMCTrans will substantially advance the development of safer imaging approaches and shorter exam-durations for pediatric patients, as well as expand the possibilities for frequent longitudinal monitoring of these patients by PET.

Deep Federated Anomaly Detection for Multivariate Time Series Data

Despite the fact that many anomaly detection approaches have been developed for multivariate time series data, limited effort has been made on federated settings in which multivariate time series data are heterogeneously distributed among different edge devices while data sharing is prohibited. In this paper, we investigate the problem of federated unsupervised anomaly detection and present a Federated Exemplar-based Deep Neural Network (Fed-ExDNN) to conduct anomaly detection for multivariate time series data on different edge devices. Specifically, we first design an Exemplar-based Deep Neural network (ExDNN) to learn local time series representations based on their compatibility with an exemplar module which consists of hidden parameters learned to capture varieties of normal patterns on each edge device. Next, a constrained clustering mechanism (FedCC) is employed on the centralized server to align and aggregate the parameters of different local exemplar modules to obtain a unified global exemplar module. Finally, the global exemplar module is deployed together with a shared feature encoder to each edge device and anomaly detection is conducted by examining the compatibility of testing data to the exemplar module. Fed-ExDNN captures local normal time series patterns with ExDNN and aggregates these patterns by FedCC, and thus can handle the heterogeneous data distributed over different edge devices simultaneously. Thoroughly empirical studies on six public datasets show that ExDNN and Fed-ExDNN can outperform state-of-the-art anomaly detection algorithms and federated learning techniques.

Visualization of Decision Trees based on General Line Coordinates to Support Explainable Models

Visualization of Machine Learning (ML) models is an important part of the ML process to enhance the interpretability and prediction accuracy of the ML models. This paper proposes a new method SPC-DT to visualize the Decision Tree (DT) as interpretable models. These methods use a version of General Line Coordinates called Shifted Paired Coordinates (SPC). In SPC, each n-D point is visualized in a set of shifted pairs of 2-D Cartesian coordinates as a directed graph. The new method expands and complements the capabilities of existing methods, to visualize DT models. It shows: (1) relations between attributes, (2) individual cases relative to the DT structure, (3) data flow in the DT, (4) how tight each split is to thresholds in the DT nodes, and (5) the density of cases in parts of the n-D space. This information is important for domain experts for evaluating and improving the DT models, including avoiding overgeneralization and overfitting of models, along with their performance. The benefits of the methods are demonstrated in the case studies, using three real datasets.

Interpretable Machine Learning for Self-Service High-Risk Decision-Making

This paper contributes to interpretable machine learning via visual knowledge discovery in general line coordinates (GLC). The concepts of hyperblocks as interpretable dataset units and general line coordinates are combined to create a visual self-service machine learning model. The DSC1 and DSC2 lossless multidimensional coordinate systems are proposed. DSC1 and DSC2 can map multiple dataset attributes to a single two-dimensional (X, Y) Cartesian plane using a graph construction algorithm. The hyperblock analysis was used to determine visually appealing dataset attribute orders and to reduce line occlusion. It is shown that hyperblocks can generalize decision tree rules and a series of DSC1 or DSC2 plots can visualize a decision tree. The DSC1 and DSC2 plots were tested on benchmark datasets from the UCI ML repository. They allowed for visual classification of data. Additionally, areas of hyperblock impurity were discovered and used to establish dataset splits that highlight the upper estimate of worst-case model accuracy to guide model selection for high-risk decision-making. Major benefits of DSC1 and DSC2 is their highly interpretable nature. They allow domain experts to control or establish new machine learning models through visual pattern discovery.

Posterior Collapse of a Linear Latent Variable Model

This work identifies the existence and cause of a type of posterior collapse that frequently occurs in the Bayesian deep learning practice. For a general linear latent variable model that includes linear variational autoencoders as a special case, we precisely identify the nature of posterior collapse to be the competition between the likelihood and the regularization of the mean due to the prior. Our result also suggests that posterior collapse may be a general problem of learning for deeper architectures and deepens our understanding of Bayesian deep learning.

ResSFL: A Resistance Transfer Framework for Defending Model Inversion Attack in Split Federated Learning

This work aims to tackle Model Inversion (MI) attack on Split Federated Learning (SFL). SFL is a recent distributed training scheme where multiple clients send intermediate activations (i.e., feature map), instead of raw data, to a central server. While such a scheme helps reduce the computational load at the client end, it opens itself to reconstruction of raw data from intermediate activation by the server. Existing works on protecting SFL only consider inference and do not handle attacks during training. So we propose ResSFL, a Split Federated Learning Framework that is designed to be MI-resistant during training. It is based on deriving a resistant feature extractor via attacker-aware training, and using this extractor to initialize the client-side model prior to standard SFL training. Such a method helps in reducing the computational complexity due to use of strong inversion model in client-side adversarial training as well as vulnerability of attacks launched in early training epochs. On CIFAR-100 dataset, our proposed framework successfully mitigates MI attack on a VGG-11 model with a high reconstruction Mean-Square-Error of 0.050 compared to 0.005 obtained by the baseline system. The framework achieves 67.5% accuracy (only 1% accuracy drop) with very low computation overhead. Code is released at: https://github.com/zlijingtao/ResSFL.

Row-wise Accelerator for Vision Transformer

Following the success of the natural language processing, the transformer for vision applications has attracted significant attention in recent years due to its excellent performance. However, existing deep learning hardware accelerators for vision cannot execute this structure efficiently due to significant model architecture differences. As a result, this paper proposes the hardware accelerator for vision transformers with row-wise scheduling, which decomposes major operations in vision transformers as a single dot product primitive for a unified and efficient execution. Furthermore, by sharing weights in columns, we can reuse the data and reduce the usage of memory. The implementation with TSMC 40nm CMOS technology only requires 262K gate count and 149KB SRAM buffer for 403.2 GOPS throughput at 600MHz clock frequency.

A Real Time Super Resolution Accelerator with Tilted Layer Fusion

Deep learning based superresolution achieves high-quality results, but its heavy computational workload, large buffer, and high external memory bandwidth inhibit its usage in mobile devices. To solve the above issues, this paper proposes a real-time hardware accelerator with the tilted layer fusion method that reduces the external DRAM bandwidth by 92\% and just needs 102KB on-chip memory. The design implemented with a 40nm CMOS process achieves 1920x1080@60fps throughput with 544.3K gate count when running at 600MHz; it has higher throughput and lower area cost than previous designs.

Hardware-Robust In-RRAM-Computing for Object Detection

In-memory computing is becoming a popular architecture for deep-learning hardware accelerators recently due to its highly parallel computing, low power, and low area cost. However, in-RRAM computing (IRC) suffered from large device variation and numerous nonideal effects in hardware. Although previous approaches including these effects in model training successfully improved variation tolerance, they only considered part of the nonideal effects and relatively simple classification tasks. This paper proposes a joint hardware and software optimization strategy to design a hardware-robust IRC macro for object detection. We lower the cell current by using a low word-line voltage to enable a complete convolution calculation in one operation that minimizes the impact of nonlinear addition. We also implement ternary weight mapping and remove batch normalization for better tolerance against device variation, sense amplifier variation, and IR drop problem. An extra bias is included to overcome the limitation of the current sensing range. The proposed approach has been successfully applied to a complex object detection task with only 3.85\% mAP drop, whereas a naive design suffers catastrophic failure under these nonideal effects.

Predicting parametric spatiotemporal dynamics by multi-resolution PDE structure-preserved deep learning

Although recent advances in deep learning (DL) have shown a great promise for learning physics exhibiting complex spatiotemporal dynamics, the high training cost, unsatisfying extrapolability for long-term predictions, and poor generalizability in out-of-sample regimes significantly limit their applications in science/engineering problems. A more promising way is to leverage available physical prior and domain knowledge to develop scientific DL models, known as physics-informed deep learning (PiDL). In most existing PiDL frameworks, e.g., physics-informed neural networks, the physics prior is mainly utilized to regularize neural network training by incorporating governing equations into the loss function in a soft manner. In this work, we propose a new direction to leverage physics prior knowledge by baking the mathematical structures of governing equations into the neural network architecture design. In particular, we develop a novel PDE-preserved neural network (PPNN) for rapidly predicting parametric spatiotemporal dynamics, given the governing PDEs are (partially) known. The discretized PDE structures are preserved in PPNN as convolutional residual network (ConvResNet) blocks, which are formulated in a multi-resolution setting. This physics-inspired learning architecture design endows PPNN with excellent generalizability and long-term prediction accuracy compared to the state-of-the-art black-box ConvResNet baseline. The effectiveness and merit of the proposed methods have been demonstrated over a handful of spatiotemporal dynamical systems governed by unsteady PDEs, including reaction-diffusion, Burgers', and Navier-Stokes equations.

Methodology to Create Analysis-Naive Holdout Records as well as Train and Test Records for Machine Learning Analyses in Healthcare

It is common for researchers to holdout data from a study pool to be used for external validation as well as for future research, and the same desire is true to those using machine learning modeling research. For this discussion, the purpose of the holdout sample it is preserve data for research studies that will be analysis-naive and randomly selected from the full dataset. Analysis-naive are records that are not used for testing or training machine learning (ML) models and records that do not participate in any aspect of the current machine learning study. The methodology suggested for creating holdouts is a modification of k-fold cross validation, which takes into account randomization and efficiently allows a three-way split (holdout, test and training) as part of the method without forcing. The paper also provides a working example using set of automated functions in Python and some scenarios for applicability in healthcare.

Building Machine Translation Systems for the Next Thousand Languages

In this paper we share findings from our effort to build practical machine translation (MT) systems capable of translating across over one thousand languages. We describe results in three research domains: (i) Building clean, web-mined datasets for 1500+ languages by leveraging semi-supervised pre-training for language identification and developing data-driven filtering techniques; (ii) Developing practical MT models for under-served languages by leveraging massively multilingual models trained with supervised parallel data for over 100 high-resource languages and monolingual datasets for an additional 1000+ languages; and (iii) Studying the limitations of evaluation metrics for these languages and conducting qualitative analysis of the outputs from our MT models, highlighting several frequent error modes of these types of models. We hope that our work provides useful insights to practitioners working towards building MT systems for currently understudied languages, and highlights research directions that can complement the weaknesses of massively multilingual models in data-sparse settings.

A Structured Span Selector

Many natural language processing tasks, e.g., coreference resolution and semantic role labeling, require selecting text spans and making decisions about them. A typical approach to such tasks is to score all possible spans and greedily select spans for task-specific downstream processing. This approach, however, does not incorporate any inductive bias about what sort of spans ought to be selected, e.g., that selected spans tend to be syntactic constituents. In this paper, we propose a novel grammar-based structured span selection model which learns to make use of the partial span-level annotation provided for such problems. Compared to previous approaches, our approach gets rid of the heuristic greedy span selection scheme, allowing us to model the downstream task on an optimal set of spans. We evaluate our model on two popular span prediction tasks: coreference resolution and semantic role labeling; and show improvements on both.

SELF-CARE: Selective Fusion with Context-Aware Low-Power Edge Computing for Stress Detection

Detecting human stress levels and emotional states with physiological body-worn sensors is a complex task, but one with many health-related benefits. Robustness to sensor measurement noise and energy efficiency of low-power devices remain key challenges in stress detection. We propose SELFCARE, a fully wrist-based method for stress detection that employs context-aware selective sensor fusion that dynamically adapts based on data from the sensors. Our method uses motion to determine the context of the system and learns to adjust the fused sensors accordingly, improving performance while maintaining energy efficiency. SELF-CARE obtains state-of-the-art performance across the publicly available WESAD dataset, achieving 86.34% and 94.12% accuracy for the 3-class and 2-class classification problems, respectively. Evaluation on real hardware shows that our approach achieves up to 2.2x (3-class) and 2.7x (2-class) energy efficiency compared to traditional sensor fusion.

Robust (Controlled) Table-to-Text Generation with Structure-Aware Equivariance Learning

Controlled table-to-text generation seeks to generate natural language descriptions for highlighted subparts of a table. Previous SOTA systems still employ a sequence-to-sequence generation method, which merely captures the table as a linear structure and is brittle when table layouts change. We seek to go beyond this paradigm by (1) effectively expressing the relations of content pieces in the table, and (2) making our model robust to content-invariant structural transformations. Accordingly, we propose an equivariance learning framework, which encodes tables with a structure-aware self-attention mechanism. This prunes the full self-attention structure into an order-invariant graph attention that captures the connected graph structure of cells belonging to the same row or column, and it differentiates between relevant cells and irrelevant cells from the structural perspective. Our framework also modifies the positional encoding mechanism to preserve the relative position of tokens in the same cell but enforce position invariance among different cells. Our technology is free to be plugged into existing table-to-text generation models, and has improved T5-based models to offer better performance on ToTTo and HiTab. Moreover, on a harder version of ToTTo, we preserve promising performance, while previous SOTA systems, even with transformation-based data augmentation, have seen significant performance drops. Our code is available at https://github.com/luka-group/Lattice.

$α$NAS: Neural Architecture Search using Property Guided Synthesis

In the past few years, neural architecture search (NAS) has become an increasingly important tool within the deep learning community. Despite the many recent successes of NAS, however, most existing approaches operate within highly structured design spaces, and hence explore only a small fraction of the full search space of neural architectures while also requiring significant manual effort from domain experts. In this work, we develop techniques that enable efficient NAS in a significantly larger design space. To accomplish this, we propose to perform NAS in an abstract search space of program properties. Our key insights are as follows: (1) the abstract search space is significantly smaller than the original search space, and (2) architectures with similar program properties also have similar performance; thus, we can search more efficiently in the abstract search space. To enable this approach, we also propose a novel efficient synthesis procedure, which accepts a set of promising program properties, and returns a satisfying neural architecture. We implement our approach, $\alpha$NAS, within an evolutionary framework, where the mutations are guided by the program properties. Starting with a ResNet-34 model, $\alpha$NAS produces a model with slightly improved accuracy on CIFAR-10 but 96% fewer parameters. On ImageNet, $\alpha$NAS is able to improve over Vision Transformer (30% fewer FLOPS and parameters), ResNet-50 (23% fewer FLOPS, 14% fewer parameters), and EfficientNet (7% fewer FLOPS and parameters) without any degradation in accuracy.

Ensemble Classifier Design Tuned to Dataset Characteristics for Network Intrusion Detection

Machine Learning-based supervised approaches require highly customized and fine-tuned methodologies to deliver outstanding performance. This paper presents a dataset-driven design and performance evaluation of a machine learning classifier for the network intrusion dataset UNSW-NB15. Analysis of the dataset suggests that it suffers from class representation imbalance and class overlap in the feature space. We employed ensemble methods using Balanced Bagging (BB), eXtreme Gradient Boosting (XGBoost), and Random Forest empowered by Hellinger Distance Decision Tree (RF-HDDT). BB and XGBoost are tuned to handle the imbalanced data, and Random Forest (RF) classifier is supplemented by the Hellinger metric to address the imbalance issue. Two new algorithms are proposed to address the class overlap issue in the dataset. These two algorithms are leveraged to help improve the performance of the testing dataset by modifying the final classification decision made by three base classifiers as part of the ensemble classifier which employs a majority vote combiner. The proposed design is evaluated for both binary and multi-category classification. Comparing the proposed model to those reported on the same dataset in the literature demonstrate that the proposed model outperforms others by a significant margin for both binary and multi-category classification cases.

Learning to Brachiate via Simplified Model Imitation

Brachiation is the primary form of locomotion for gibbons and siamangs, in which these primates swing from tree limb to tree limb using only their arms. It is challenging to control because of the limited control authority, the required advance planning, and the precision of the required grasps. We present a novel approach to this problem using reinforcement learning, and as demonstrated on a finger-less 14-link planar model that learns to brachiate across challenging handhold sequences. Key to our method is the use of a simplified model, a point mass with a virtual arm, for which we first learn a policy that can brachiate across handhold sequences with a prescribed order. This facilitates the learning of the policy for the full model, for which it provides guidance by providing an overall center-of-mass trajectory to imitate, as well as for the timing of the holds. Lastly, the simplified model can also readily be used for planning suitable sequences of handholds in a given environment. Our results demonstrate brachiation motions with a variety of durations for the flight and hold phases, as well as emergent extra back-and-forth swings when this proves useful. The system is evaluated with a variety of ablations. The method enables future work towards more general 3D brachiation, as well as using simplified model imitation in other settings.

Investigating Generalization by Controlling Normalized Margin

Weight norm $\|w\|$ and margin $\gamma$ participate in learning theory via the normalized margin $\gamma/\|w\|$. Since standard neural net optimizers do not control normalized margin, it is hard to test whether this quantity causally relates to generalization. This paper designs a series of experimental studies that explicitly control normalized margin and thereby tackle two central questions. First: does normalized margin always have a causal effect on generalization? The paper finds that no -- networks can be produced where normalized margin has seemingly no relationship with generalization, counter to the theory of Bartlett et al. (2017). Second: does normalized margin ever have a causal effect on generalization? The paper finds that yes -- in a standard training setup, test performance closely tracks normalized margin. The paper suggests a Gaussian process model as a promising explanation for this behavior.

N-ACT: An Interpretable Deep Learning Model for Automatic Cell Type and Salient Gene Identification

Single-cell RNA sequencing (scRNAseq) is rapidly advancing our understanding of cellular composition within complex tissues and organisms. A major limitation in most scRNAseq analysis pipelines is the reliance on manual annotations to determine cell identities, which are time consuming, subjective, and require expertise. Given the surge in cell sequencing, supervised methods-especially deep learning models-have been developed for automatic cell type identification (ACTI), which achieve high accuracy and scalability. However, all existing deep learning frameworks for ACTI lack interpretability and are used as "black-box" models. We present N-ACT (Neural-Attention for Cell Type identification): the first-of-its-kind interpretable deep neural network for ACTI utilizing neural-attention to detect salient genes for use in cell-type identification. We compare N-ACT to conventional annotation methods on two previously manually annotated data sets, demonstrating that N-ACT accurately identifies marker genes and cell types in an unsupervised manner, while performing comparably on multiple data sets to current state-of-the-art model in traditional supervised ACTI.

Unsupervised Discovery and Composition of Object Light Fields

Neural scene representations, both continuous and discrete, have recently emerged as a powerful new paradigm for 3D scene understanding. Recent efforts have tackled unsupervised discovery of object-centric neural scene representations. However, the high cost of ray-marching, exacerbated by the fact that each object representation has to be ray-marched separately, leads to insufficiently sampled radiance fields and thus, noisy renderings, poor framerates, and high memory and time complexity during training and rendering. Here, we propose to represent objects in an object-centric, compositional scene representation as light fields. We propose a novel light field compositor module that enables reconstructing the global light field from a set of object-centric light fields. Dubbed Compositional Object Light Fields (COLF), our method enables unsupervised learning of object-centric neural scene representations, state-of-the-art reconstruction and novel view synthesis performance on standard datasets, and rendering and training speeds at orders of magnitude faster than existing 3D approaches.

Online Algorithms with Multiple Predictions

This paper studies online algorithms augmented with multiple machine-learned predictions. While online algorithms augmented with a single prediction have been extensively studied in recent years, the literature for the multiple predictions setting is sparse. In this paper, we give a generic algorithmic framework for online covering problems with multiple predictions that obtains an online solution that is competitive against the performance of the best predictor. Our algorithm incorporates the use of predictions in the classic potential-based analysis of online algorithms. We apply our algorithmic framework to solve classical problems such as online set cover, (weighted) caching, and online facility location in the multiple predictions setting. Our algorithm can also be robustified, i.e., the algorithm can be simultaneously made competitive against the best prediction and the performance of the best online algorithm (without prediction).

Federated Random Reshuffling with Compression and Variance Reduction

Random Reshuffling (RR), which is a variant of Stochastic Gradient Descent (SGD) employing sampling without replacement, is an immensely popular method for training supervised machine learning models via empirical risk minimization. Due to its superior practical performance, it is embedded and often set as default in standard machine learning software. Under the name FedRR, this method was recently shown to be applicable to federated learning (Mishchenko et al.,2021), with superior performance when compared to common baselines such as Local SGD. Inspired by this development, we design three new algorithms to improve FedRR further: compressed FedRR and two variance reduced extensions: one for taming the variance coming from shuffling and the other for taming the variance due to compression. The variance reduction mechanism for compression allows us to eliminate dependence on the compression parameter, and applying additional controlled linear perturbations for Random Reshuffling, introduced by Malinovsky et al.(2021) helps to eliminate variance at the optimum. We provide the first analysis of compressed local methods under standard assumptions without bounded gradient assumptions and for heterogeneous data, overcoming the limitations of the compression operator. We corroborate our theoretical results with experiments on synthetic and real data sets.

Dynamic categories, dynamic operads: From deep learning to prediction markets

Natural organized systems adapt to internal and external pressures and this seems to happens all the way down. Wanting to think clearly about this idea motivates our paper, and so the idea is elaborated extensively in the introduction, which should be broadly accessible to a philosophically-interested audience. In the remaining sections, we turn to more compressed category theory. We define the monoidal double category $\mathbf{Org}$ of dynamic organizations, we provide definitions of $\mathbf{Org}$-enriched, or "dynamic", categorical structures -- e.g. dynamic categories, operads, and monoidal categories -- and we show how they instantiate the motivating philosophical ideas. We give two examples of dynamic categorical structures: prediction markets as a dynamic operad and deep learning as a dynamic monoidal category.

Decentralized Stochastic Optimization with Inherent Privacy Protection

Decentralized stochastic optimization is the basic building block of modern collaborative machine learning, distributed estimation and control, and large-scale sensing. Since involved data usually contain sensitive information like user locations, healthcare records and financial transactions, privacy protection has become an increasingly pressing need in the implementation of decentralized stochastic optimization algorithms. In this paper, we propose a decentralized stochastic gradient descent algorithm which is embedded with inherent privacy protection for every participating agent against other participating agents and external eavesdroppers. This proposed algorithm builds in a dynamics based gradient-obfuscation mechanism to enable privacy protection without compromising optimization accuracy, which is in significant difference from differential-privacy based privacy solutions for decentralized optimization that have to trade optimization accuracy for privacy. The dynamics based privacy approach is encryption-free, and hence avoids incurring heavy communication or computation overhead, which is a common problem with encryption based privacy solutions for decentralized stochastic optimization. Besides rigorously characterizing the convergence performance of the proposed decentralized stochastic gradient descent algorithm under both convex objective functions and non-convex objective functions, we also provide rigorous information-theoretic analysis of its strength of privacy protection. Simulation results for a distributed estimation problem as well as numerical experiments for decentralized learning on a benchmark machine learning dataset confirm the effectiveness of the proposed approach.

Multimodal Semi-Supervised Learning for Text Recognition

Until recently, the number of public real-world text images was insufficient for training scene text recognizers. Therefore, most modern training methods rely on synthetic data and operate in a fully supervised manner. Nevertheless, the amount of public real-world text images has increased significantly lately, including a great deal of unlabeled data. Leveraging these resources requires semi-supervised approaches; however, the few existing methods do not account for vision-language multimodality structure and therefore suboptimal for state-of-the-art multimodal architectures. To bridge this gap, we present semi-supervised learning for multimodal text recognizers (SemiMTR) that leverages unlabeled data at each modality training phase. Notably, our method refrains from extra training stages and maintains the current three-stage multimodal training procedure. Our algorithm starts by pretraining the vision model through a single-stage training that unifies self-supervised learning with supervised training. More specifically, we extend an existing visual representation learning algorithm and propose the first contrastive-based method for scene text recognition. After pretraining the language model on a text corpus, we fine-tune the entire network via a sequential, character-level, consistency regularization between weakly and strongly augmented views of text images. In a novel setup, consistency is enforced on each modality separately. Extensive experiments validate that our method outperforms the current training schemes and achieves state-of-the-art results on multiple scene text recognition benchmarks.

Neural Program Synthesis with Query

Aiming to find a program satisfying the user intent given input-output examples, program synthesis has attracted increasing interest in the area of machine learning. Despite the promising performance of existing methods, most of their success comes from the privileged information of well-designed input-output examples. However, providing such input-output examples is unrealistic because it requires the users to have the ability to describe the underlying program with a few input-output examples under the training distribution. In this work, we propose a query-based framework that trains a query neural network to generate informative input-output examples automatically and interactively from a large query space. The quality of the query depends on the amount of the mutual information between the query and the corresponding program, which can guide the optimization of the query framework. To estimate the mutual information more accurately, we introduce the functional space (F-space) which models the relevance between the input-output examples and the programs in a differentiable way. We evaluate the effectiveness and generalization of the proposed query-based framework on the Karel task and the list processing task. Experimental results show that the query-based framework can generate informative input-output examples which achieve and even outperform well-designed input-output examples.

On Conditioning the Input Noise for Controlled Image Generation with Diffusion Models

Conditional image generation has paved the way for several breakthroughs in image editing, generating stock photos and 3-D object generation. This continues to be a significant area of interest with the rise of new state-of-the-art methods that are based on diffusion models. However, diffusion models provide very little control over the generated image, which led to subsequent works exploring techniques like classifier guidance, that provides a way to trade off diversity with fidelity. In this work, we explore techniques to condition diffusion models with carefully crafted input noise artifacts. This allows generation of images conditioned on semantic attributes. This is different from existing approaches that input Gaussian noise and further introduce conditioning at the diffusion model's inference step. Our experiments over several examples and conditional settings show the potential of our approach.

Assigning Species Information to Corresponding Genes by a Sequence Labeling Framework

The automatic assignment of species information to the corresponding genes in a research article is a critically important step in the gene normalization task, whereby a gene mention is normalized and linked to a database record or identifier by a text-mining algorithm. Existing methods typically rely on heuristic rules based on gene and species co-occurrence in the article, but their accuracy is suboptimal. We therefore developed a high-performance method, using a novel deep learning-based framework, to classify whether there is a relation between a gene and a species. Instead of the traditional binary classification framework in which all possible pairs of genes and species in the same article are evaluated, we treat the problem as a sequence-labeling task such that only a fraction of the pairs needs to be considered. Our benchmarking results show that our approach obtains significantly higher performance compared to that of the rule-based baseline method for the species assignment task (from 65.8% to 81.3% in accuracy). The source code and data for species assignment are freely available at https://github.com/ncbi/SpeciesAssignment.

Differentiable Electron Microscopy Simulation: Methods and Applications for Visualization

We propose a new microscopy simulation system that can depict atomistic models in a micrograph visual style, similar to results of physical electron microscopy imaging. This system is scalable, able to represent simulation of electron microscopy of tens of viral particles and synthesizes the image faster than previous methods. On top of that, the simulator is differentiable, both its deterministic as well as stochastic stages that form signal and noise representations in the micrograph. This notable property has the capability for solving inverse problems by means of optimization and thus allows for generation of microscopy simulations using the parameter settings estimated from real data. We demonstrate this learning capability through two applications: (1) estimating the parameters of the modulation transfer function defining the detector properties of the simulated and real micrographs, and (2) denoising the real data based on parameters trained from the simulated examples. While current simulators do not support any parameter estimation due to their forward design, we show that the results obtained using estimated parameters are very similar to the results of real micrographs. Additionally, we evaluate the denoising capabilities of our approach and show that the results showed an improvement over state-of-the-art methods. Denoised micrographs exhibit less noise in the tilt-series tomography reconstructions, ultimately reducing the visual dominance of noise in direct volume rendering of microscopy tomograms.

SeqNet: An Efficient Neural Network for Automatic Malware Detection

Malware continues to evolve rapidly, and more than 450,000 new samples are captured every day, which makes manual malware analysis impractical. However, existing deep learning detection models need manual feature engineering or require high computational overhead for long training processes, which might be laborious to select feature space and difficult to retrain for mitigating model aging. Therefore, a crucial requirement for a detector is to realize automatic and efficient detection. In this paper, we propose a lightweight malware detection model called SeqNet which could be trained at high speed with low memory required on the raw binaries. By avoiding contextual confusion and reducing semantic loss, SeqNet maintains the detection accuracy when reducing the number of parameters to only 136K. We demonstrate the effectiveness of our methods and the low training cost requirement of SeqNet in our experiments. Besides, we make our datasets and codes public to stimulate further academic research.

FP-GNN: a versatile deep learning architecture for enhanced molecular property prediction

Deep learning is an important method for molecular design and exhibits considerable ability to predict molecular properties, including physicochemical, bioactive, and ADME/T (absorption, distribution, metabolism, excretion, and toxicity) properties. In this study, we advanced a novel deep learning architecture, termed FP-GNN, which combined and simultaneously learned information from molecular graphs and fingerprints. To evaluate the FP-GNN model, we conducted experiments on 13 public datasets, an unbiased LIT-PCBA dataset, and 14 phenotypic screening datasets for breast cell lines. Extensive evaluation results showed that compared to advanced deep learning and conventional machine learning algorithms, the FP-GNN algorithm achieved state-of-the-art performance on these datasets. In addition, we analyzed the influence of different molecular fingerprints, and the effects of molecular graphs and molecular fingerprints on the performance of the FP-GNN model. Analysis of the anti-noise ability and interpretation ability also indicated that FP-GNN was competitive in real-world situations.

Accelerated functional brain aging in major depressive disorder: evidence from a large scale fMRI analysis of Chinese participants

Major depressive disorder (MDD) is one of the most common mental health conditions that has been intensively investigated for its association with brain atrophy and mortality. Recent studies reveal that the deviation between the predicted and the chronological age can be a marker of accelerated brain aging to characterize MDD. However, current conclusions are usually drawn based on structural MRI information collected from Caucasian participants. The universality of this biomarker needs to be further validated by subjects with different ethnic/racial backgrounds and by different types of data. Here we make use of the REST-meta-MDD, a large scale resting-state fMRI dataset collected from multiple cohort participants in China. We develop a stacking machine learning model based on 1101 healthy controls, which estimates a subject's chronological age from fMRI with promising accuracy. The trained model is then applied to 1276 MDD patients from 24 sites. We observe that MDD patients exhibit a $+4.43$ years ($\text{$p$} < 0.0001$, $\text{Cohen's $d$} = 0.35$, $\text{95\% CI}:1.86 - 3.91$) higher brain-predicted age difference (brain-PAD) compared to controls. In the MDD subgroup, we observe a statistically significant $+2.09$ years ($\text{$p$} < 0.05$, $\text{Cohen's $d$} = 0.134483$) brain-PAD in antidepressant users compared to medication-free patients. The statistical relationship observed is further checked by three different machine learning algorithms. The positive brain-PAD observed in participants in China confirms the presence of accelerated brain aging in MDD patients. The utilization of functional brain connectivity for age estimation verifies existing findings from a new dimension.

Some performance considerations when using multi-armed bandit algorithms in the presence of missing data

When using multi-armed bandit algorithms, the potential impact of missing data is often overlooked. In practice, the simplest approach is to ignore missing outcomes and continue to sample following the bandit algorithm. We investigate the impact of missing data on several bandit algorithms via a simulation study assuming the rewards are missing at random. We focus on two-armed bandit algorithms with binary outcomes in the context of patient allocation for clinical trials with relatively small sample sizes. However, our results can apply to other applications of bandit algorithms where missing data is expected to occur. We assess the resulting operating characteristics, including the expected reward (i.e., allocation results). Different probabilities of missingness in both arms are considered. The key finding of our work is that when using the simplest strategy of ignoring missing data, the corresponding impact on the performance of multi-armed bandit strategies varies according to their way of balancing the exploration-exploitation trade-off. Algorithms that are geared towards exploration continue to assign samples to the arm with more missing responses, and this arm is perceived as the superior arm by the algorithm. By contrast, algorithms that are geared towards exploitation would do the opposite and not assign samples to the arms with more missing responses. Furthermore, for algorithms focusing more on exploration, we illustrate that the problem of missing responses can be alleviated using a simple mean imputation approach.

Simultaneous Double Q-learning with Conservative Advantage Learning for Actor-Critic Methods

Actor-critic Reinforcement Learning (RL) algorithms have achieved impressive performance in continuous control tasks. However, they still suffer two nontrivial obstacles, i.e., low sample efficiency and overestimation bias. To this end, we propose Simultaneous Double Q-learning with Conservative Advantage Learning (SDQ-CAL). Our SDQ-CAL boosts the Double Q-learning for off-policy actor-critic RL based on a modification of the Bellman optimality operator with Advantage Learning. Specifically, SDQ-CAL improves sample efficiency by modifying the reward to facilitate the distinction from experience between the optimal actions and the others. Besides, it mitigates the overestimation issue by updating a pair of critics simultaneously upon double estimators. Extensive experiments reveal that our algorithm realizes less biased value estimation and achieves state-of-the-art performance in a range of continuous control benchmark tasks. We release the source code of our method at: \url{https://github.com/LQNew/SDQ-CAL}.

Data-Free Adversarial Knowledge Distillation for Graph Neural Networks

Graph neural networks (GNNs) have been widely used in modeling graph structured data, owing to its impressive performance in a wide range of practical applications. Recently, knowledge distillation (KD) for GNNs has enabled remarkable progress in graph model compression and knowledge transfer. However, most of the existing KD methods require a large volume of real data, which are not readily available in practice, and may preclude their applicability in scenarios where the teacher model is trained on rare or hard to acquire datasets. To address this problem, we propose the first end-to-end framework for data-free adversarial knowledge distillation on graph structured data (DFAD-GNN). To be specific, our DFAD-GNN employs a generative adversarial network, which mainly consists of three components: a pre-trained teacher model and a student model are regarded as two discriminators, and a generator is utilized for deriving training graphs to distill knowledge from the teacher model into the student model. Extensive experiments on various benchmark models and six representative datasets demonstrate that our DFAD-GNN significantly surpasses state-of-the-art data-free baselines in the graph classification task.

Over-the-Air Federated Multi-Task Learning via Model Sparsification and Turbo Compressed Sensing

To achieve communication-efficient federated multitask learning (FMTL), we propose an over-the-air FMTL (OAFMTL) framework, where multiple learning tasks deployed on edge devices share a non-orthogonal fading channel under the coordination of an edge server (ES). In OA-FMTL, the local updates of edge devices are sparsified, compressed, and then sent over the uplink channel in a superimposed fashion. The ES employs over-the-air computation in the presence of intertask interference. More specifically, the model aggregations of all the tasks are reconstructed from the channel observations concurrently, based on a modified version of the turbo compressed sensing (Turbo-CS) algorithm (named as M-Turbo-CS). We analyze the performance of the proposed OA-FMTL framework together with the M-Turbo-CS algorithm. Furthermore, based on the analysis, we formulate a communication-learning optimization problem to improve the system performance by adjusting the power allocation among the tasks at the edge devices. Numerical simulations show that our proposed OAFMTL effectively suppresses the inter-task interference, and achieves a learning performance comparable to its counterpart with orthogonal multi-task transmission. It is also shown that the proposed inter-task power allocation optimization algorithm substantially reduces the overall communication overhead by appropriately adjusting the power allocation among the tasks.

Deep Embedded Multi-View Clustering via Jointly Learning Latent Representations and Graphs

With the representation learning capability of the deep learning models, deep embedded multi-view clustering (MVC) achieves impressive performance in many scenarios and has become increasingly popular in recent years. Although great progress has been made in this field, most existing methods merely focus on learning the latent representations and ignore that learning the latent graph of nodes also provides available information for the clustering task. To address this issue, in this paper we propose Deep Embedded Multi-view Clustering via Jointly Learning Latent Representations and Graphs (DMVCJ), which utilizes the latent graphs to promote the performance of deep embedded MVC models from two aspects. Firstly, by learning the latent graphs and feature representations jointly, the graph convolution network (GCN) technique becomes available for our model. With the capability of GCN in exploiting the information from both graphs and features, the clustering performance of our model is significantly promoted. Secondly, based on the adjacency relations of nodes shown in the latent graphs, we design a sample-weighting strategy to alleviate the noisy issue, and further improve the effectiveness and robustness of the model. Experimental results on different types of real-world multi-view datasets demonstrate the effectiveness of DMVCJ.

Univariate and Multivariate LSTM Model for Short-Term Stock Market Prediction

Designing robust and accurate prediction models has been a viable research area since a long time. While proponents of a well-functioning market predictors believe that it is difficult to accurately predict market prices but many scholars disagree. Robust and accurate prediction systems will not only be helpful to the businesses but also to the individuals in making their financial investments. This paper presents an LSTM model with two different input approaches for predicting the short-term stock prices of two Indian companies, Reliance Industries and Infosys Ltd. Ten years of historic data (2012-2021) is taken from the yahoo finance website to carry out analysis of proposed approaches. In the first approach, closing prices of two selected companies are directly applied on univariate LSTM model. For the approach second, technical indicators values are calculated from the closing prices and then collectively applied on Multivariate LSTM model. Short term market behaviour for upcoming days is evaluated. Experimental outcomes revel that approach one is useful to determine the future trend but multivariate LSTM model with technical indicators found to be useful in accurately predicting the future price behaviours.

Pervasive Machine Learning for Smart Radio Environments Enabled by Reconfigurable Intelligent Surfaces

The emerging technology of Reconfigurable Intelligent Surfaces (RISs) is provisioned as an enabler of smart wireless environments, offering a highly scalable, low-cost, hardware-efficient, and almost energy-neutral solution for dynamic control of the propagation of electromagnetic signals over the wireless medium, ultimately providing increased environmental intelligence for diverse operation objectives. One of the major challenges with the envisioned dense deployment of RISs in such reconfigurable radio environments is the efficient configuration of multiple metasurfaces with limited, or even the absence of, computing hardware. In this paper, we consider multi-user and multi-RIS-empowered wireless systems, and present a thorough survey of the online machine learning approaches for the orchestration of their various tunable components. Focusing on the sum-rate maximization as a representative design objective, we present a comprehensive problem formulation based on Deep Reinforcement Learning (DRL). We detail the correspondences among the parameters of the wireless system and the DRL terminology, and devise generic algorithmic steps for the artificial neural network training and deployment, while discussing their implementation details. Further practical considerations for multi-RIS-empowered wireless communications in the sixth Generation (6G) era are presented along with some key open research challenges. Differently from the DRL-based status quo, we leverage the independence between the configuration of the system design parameters and the future states of the wireless environment, and present efficient multi-armed bandits approaches, whose resulting sum-rate performances are numerically shown to outperform random configurations, while being sufficiently close to the conventional Deep Q-Network (DQN) algorithm, but with lower implementation complexity.

MLSmellHound: A Context-Aware Code Analysis Tool

Meeting the rise of industry demand to incorporate machine learning (ML) components into software systems requires interdisciplinary teams contributing to a shared code base. To maintain consistency, reduce defects and ensure maintainability, developers use code analysis tools to aid them in identifying defects and maintaining standards. With the inclusion of machine learning, tools must account for the cultural differences within the teams which manifests as multiple programming languages, and conflicting definitions and objectives. Existing tools fail to identify these cultural differences and are geared towards software engineering which reduces their adoption in ML projects. In our approach we attempt to resolve this problem by exploring the use of context which includes i) purpose of the source code, ii) technical domain, iii) problem domain, iv) team norms, v) operational environment, and vi) development lifecycle stage to provide contextualised error reporting for code analysis. To demonstrate our approach, we adapt Pylint as an example and apply a set of contextual transformations to the linting results based on the domain of individual project files under analysis. This allows for contextualised and meaningful error reporting for the end-user.

Learning Regionally Decentralized AC Optimal Power Flows with ADMM

One potential future for the next generation of smart grids is the use of decentralized optimization algorithms and secured communications for coordinating renewable generation (e.g., wind/solar), dispatchable devices (e.g., coal/gas/nuclear generations), demand response, battery & storage facilities, and topology optimization. The Alternating Direction Method of Multipliers (ADMM) has been widely used in the community to address such decentralized optimization problems and, in particular, the AC Optimal Power Flow (AC-OPF). This paper studies how machine learning may help in speeding up the convergence of ADMM for solving AC-OPF. It proposes a novel decentralized machine-learning approach, namely ML-ADMM, where each agent uses deep learning to learn the consensus parameters on the coupling branches. The paper also explores the idea of learning only from ADMM runs that exhibit high-quality convergence properties, and proposes filtering mechanisms to select these runs. Experimental results on test cases based on the French system demonstrate the potential of the approach in speeding up the convergence of ADMM significantly.