Datasets:

CShorten

/

ML-ArXiv-Papers

like 45

An Exponentially Increasing Step-size for Parameter Estimation in Statistical Models

Using gradient descent (GD) with fixed or decaying step-size is standard practice in unconstrained optimization problems. However, when the loss function is only locally convex, such a step-size schedule artificially slows GD down as it cannot explore the flat curvature of the loss function. To overcome that issue, we propose to exponentially increase the step-size of the GD algorithm. Under homogeneous assumptions on the loss function, we demonstrate that the iterates of the proposed \emph{exponential step size gradient descent} (EGD) algorithm converge linearly to the optimal solution. Leveraging that optimization insight, we then consider using the EGD algorithm for solving parameter estimation under non-regular statistical models whose the loss function becomes locally convex when the sample size goes to infinity. We demonstrate that the EGD iterates reach the final statistical radius within the true parameter after a logarithmic number of iterations, which is in stark contrast to a \emph{polynomial} number of iterations of the GD algorithm. Therefore, the total computational complexity of the EGD algorithm is \emph{optimal} and exponentially cheaper than that of the GD for solving parameter estimation in non-regular statistical models. To the best of our knowledge, it resolves a long-standing gap between statistical and algorithmic computational complexities of parameter estimation in non-regular statistical models. Finally, we provide targeted applications of the general theory to several classes of statistical models, including generalized linear models with polynomial link functions and location Gaussian mixture models.

Deep Apprenticeship Learning for Playing Games

In the last decade, deep learning has achieved great success in machine learning tasks where the input data is represented with different levels of abstractions. Driven by the recent research in reinforcement learning using deep neural networks, we explore the feasibility of designing a learning model based on expert behaviour for complex, multidimensional tasks where reward function is not available. We propose a novel method for apprenticeship learning based on the previous research on supervised learning techniques in reinforcement learning. Our method is applied to video frames from Atari games in order to teach an artificial agent to play those games. Even though the reported results are not comparable with the state-of-the-art results in reinforcement learning, we demonstrate that such an approach has the potential to achieve strong performance in the future and is worthwhile for further research.

Application of multilayer perceptron with data augmentation in nuclear physics

Neural networks have become popular in many fields of science since they serve as reliable and powerful tools. Application of the neural networks to the nuclear physics studies has also become popular in recent years because of their success in the prediction of nuclear properties. In this work, we study the effect of the data augmentation on the predictive power of the neural network models. Even though there are various data augmentation techniques used for classification tasks in the literature, this area is still very limited for regression problems. As predicting the binding energies is statistically defined as a regression problem, in addition to using data augmentation for nuclear physics, this study contributes to this field for regression in general. Using the experimental uncertainties for data augmentation, the size of training data set is artificially boosted and the changes in the root-mean-square error between the model predictions on test set and the experimental data are investigated. As far as we know, this is the first time that data augmentation techniques have been implemented for nuclear physics research. Our results show that the data augmentation decreases the prediction errors, stabilizes the model and prevents overfitting. The extrapolation capabilities of the MLP models with different depths are also tested for newly measured nuclei in AME2020 mass table.

Neural-Symbolic Models for Logical Queries on Knowledge Graphs

Answering complex first-order logic (FOL) queries on knowledge graphs is a fundamental task for multi-hop reasoning. Traditional symbolic methods traverse a complete knowledge graph to extract the answers, which provides good interpretation for each step. Recent neural methods learn geometric embeddings for complex queries. These methods can generalize to incomplete knowledge graphs, but their reasoning process is hard to interpret. In this paper, we propose Graph Neural Network Query Executor (GNN-QE), a neural-symbolic model that enjoys the advantages of both worlds. GNN-QE decomposes a complex FOL query into relation projections and logical operations over fuzzy sets, which provides interpretability for intermediate variables. To reason about the missing links, GNN-QE adapts a graph neural network from knowledge graph completion to execute the relation projections, and models the logical operations with product fuzzy logic. Extensive experiments on 3 datasets show that GNN-QE significantly improves over previous state-of-the-art models in answering FOL queries. Meanwhile, GNN-QE can predict the number of answers without explicit supervision, and provide visualizations for intermediate variables.

Distributed Feature Selection for High-dimensional Additive Models

Distributed statistical learning is a common strategy for handling massive data where we divide the learning task into multiple local machines and aggregate the results afterward. However, most existing work considers the case where the samples are divided. In this work, we propose a new algorithm, DDAC-SpAM, that divides features under the high-dimensional sparse additive model. The new algorithm contains three steps: divide, decorrelate, and conquer. We show that after the decorrelation operation, every local estimator can recover the sparsity pattern for each additive component consistently without imposing strict constraints to the correlation structure among variables. Theoretical analysis of the aggregated estimator and empirical results on synthetic and real data illustrate that the DDAC-SpAM algorithm is effective and competitive in fitting sparse additive models.

An Extension to Basis-Hypervectors for Learning from Circular Data in Hyperdimensional Computing

Hyperdimensional Computing (HDC) is a computation framework based on properties of high-dimensional random spaces. It is particularly useful for machine learning in resource-constrained environments, such as embedded systems and IoT, as it achieves a good balance between accuracy, efficiency and robustness. The mapping of information to the hyperspace, named encoding, is the most important stage in HDC. At its heart are basis-hypervectors, responsible for representing the smallest units of meaningful information. In this work we present a detailed study on basis-hypervector sets, which leads to practical contributions to HDC in general: 1) we propose an improvement for level-hypervectors, used to encode real numbers; 2) we introduce a method to learn from circular data, an important type of information never before addressed in machine learning with HDC. Empirical results indicate that these contributions lead to considerably more accurate models for both classification and regression with circular data.

Does Crypto Kill? Relationship between Electricity Consumption Carbon Footprints and Bitcoin Transactions

Cryptocurrencies are gaining more popularity due to their security, making counterfeits impossible. However, these digital currencies have been criticized for creating a large carbon footprint due to their algorithmic complexity and decentralized system design for proof of work and mining. We hypothesize that the carbon footprint of cryptocurrency transactions has a higher dependency on carbon-rich fuel sources than green or renewable fuel sources. We provide a machine learning framework to model such transactions and correlate them with the electricity generation patterns to estimate and analyze their carbon cost.

Fat-Tailed Variational Inference with Anisotropic Tail Adaptive Flows

While fat-tailed densities commonly arise as posterior and marginal distributions in robust models and scale mixtures, they present challenges when Gaussian-based variational inference fails to capture tail decay accurately. We first improve previous theory on tails of Lipschitz flows by quantifying how the tails affect the rate of tail decay and by expanding the theory to non-Lipschitz polynomial flows. Then, we develop an alternative theory for multivariate tail parameters which is sensitive to tail-anisotropy. In doing so, we unveil a fundamental problem which plagues many existing flow-based methods: they can only model tail-isotropic distributions (i.e., distributions having the same tail parameter in every direction). To mitigate this and enable modeling of tail-anisotropic targets, we propose anisotropic tail-adaptive flows (ATAF). Experimental results on both synthetic and real-world targets confirm that ATAF is competitive with prior work while also exhibiting appropriate tail-anisotropy.

Fast and realistic large-scale structure from machine-learning-augmented random field simulations

Producing thousands of simulations of the dark matter distribution in the Universe with increasing precision is a challenging but critical task to facilitate the exploitation of current and forthcoming cosmological surveys. Many inexpensive substitutes to full $N$-body simulations have been proposed, even though they often fail to reproduce the statistics of the smaller, non-linear scales. Among these alternatives, a common approximation is represented by the lognormal distribution, which comes with its own limitations as well, while being extremely fast to compute even for high-resolution density fields. In this work, we train a machine learning model to transform projected lognormal dark matter density fields to more realistic dark matter maps, as obtained from full $N$-body simulations. We detail the procedure that we follow to generate highly correlated pairs of lognormal and simulated maps, which we use as our training data, exploiting the information of the Fourier phases. We demonstrate the performance of our model comparing various statistical tests with different field resolutions, redshifts and cosmological parameters, proving its robustness and explaining its current limitations. The augmented lognormal random fields reproduce the power spectrum up to wavenumbers of $1 \ h \ \rm{Mpc}^{-1}$, the bispectrum and the peak counts within 10%, and always within the error bars, of the fiducial target simulations. Finally, we describe how we plan to integrate our proposed model with existing tools to yield more accurate spherical random fields for weak lensing analysis, going beyond the lognormal approximation.

Power and limitations of single-qubit native quantum neural networks

Quantum neural networks (QNNs) have emerged as a leading strategy to establish applications in machine learning, chemistry, and optimization. While the applications of QNN have been widely investigated, its theoretical foundation remains less understood. In this paper, we formulate a theoretical framework for the expressive ability of data re-uploading quantum neural networks that consist of interleaved encoding circuit blocks and trainable circuit blocks. First, we prove that single-qubit quantum neural networks can approximate any univariate function by mapping the model to a partial Fourier series. Beyond previous works' understanding of existence, we in particular establish the exact correlations between the parameters of the trainable gates and the working Fourier coefficients, by exploring connections to quantum signal processing. Second, we discuss the limitations of single-qubit native QNNs on approximating multivariate functions by analyzing the frequency spectrum and the flexibility of Fourier coefficients. We further demonstrate the expressivity and limitations of single-qubit native QNNs via numerical experiments. As applications, we introduce natural extensions to multi-qubit quantum neural networks, which exhibit the capability of classifying real-world multi-dimensional data. We believe these results would improve our understanding of QNNs and provide a helpful guideline for designing powerful QNNs for machine learning tasks.

Loss Landscape Engineering via Data Regulation on PINNs

Physics-Informed Neural Networks have shown unique utility in parameterising the solution of a well-defined partial differential equation using automatic differentiation and residual losses. Though they provide theoretical guarantees of convergence, in practice the required training regimes tend to be exacting and demanding. Through the course of this paper, we take a deep dive into understanding the loss landscapes associated with a PINN and how that offers some insight as to why PINNs are fundamentally hard to optimise for. We demonstrate how PINNs can be forced to converge better towards the solution, by way of feeding in sparse or coarse data as a regulator. The data regulates and morphs the topology of the loss landscape associated with the PINN to make it easily traversable for the minimiser. Data regulation of PINNs helps ease the optimisation required for convergence by invoking a hybrid unsupervised-supervised training approach, where the labelled data pushes the network towards the vicinity of the solution, and the unlabelled regime fine-tunes it to the solution.

Physics-informed machine learning techniques for edge plasma turbulence modelling in computational theory and experiment

Edge plasma turbulence is critical to the performance of magnetic confinement fusion devices. Towards better understanding edge turbulence in both theory and experiment, a custom-built physics-informed deep learning framework constrained by partial differential equations is developed to accurately learn turbulent fields consistent with the two-fluid theory from partial observations of electron pressure. This calculation is not otherwise possible using conventional equilibrium models. With this technique, the first direct quantitative comparisons of turbulent fields between electrostatic two-fluid theory and electromagnetic gyrokinetic modelling are demonstrated with good overall agreement found in magnetized helical plasmas at low normalized pressure. To translate these computational techniques to experimental fusion plasmas, a novel method to translate brightness measurements of HeI line radiation into local plasma fluctuations is demonstrated via a newly created deep learning framework that integrates neutral transport physics and collisional radiative theory for the $3^3 D - 2^3 P$ transition in atomic helium. Using fast camera data on the Alcator C-Mod tokamak, this thesis presents the first 2-dimensional time-dependent experimental measurements of the turbulent electron density, electron temperature, and neutral density in a fusion plasma using a single spectral line. With this experimentally inferred data, initial estimates of the 2-dimensional turbulent electric field consistent with drift-reduced Braginskii theory under the framework of an axisymmetric fusion plasma with purely toroidal field are calculated. The inclusion of atomic helium effects on particle and energy sources are found to strengthen correlations between the electric field and electron pressure while broadening turbulent field amplitudes which impact ${\bf E \times B}$ flows and shearing rates.

Decision Making for Hierarchical Multi-label Classification with Multidimensional Local Precision Rate

Hierarchical multi-label classification (HMC) has drawn increasing attention in the past few decades. It is applicable when hierarchical relationships among classes are available and need to be incorporated along with the multi-label classification whereby each object is assigned to one or more classes. There are two key challenges in HMC: i) optimizing the classification accuracy, and meanwhile ii) ensuring the given class hierarchy. To address these challenges, in this article, we introduce a new statistic called the multidimensional local precision rate (mLPR) for each object in each class. We show that classification decisions made by simply sorting objects across classes in descending order of their true mLPRs can, in theory, ensure the class hierarchy and lead to the maximization of CATCH, an objective function we introduce that is related to the area under a hit curve. This approach is the first of its kind that handles both challenges in one objective function without additional constraints, thanks to the desirable statistical properties of CATCH and mLPR. In practice, however, true mLPRs are not available. In response, we introduce HierRank, a new algorithm that maximizes an empirical version of CATCH using estimated mLPRs while respecting the hierarchy. The performance of this approach was evaluated on a synthetic data set and two real data sets; ours was found to be superior to several comparison methods on evaluation criteria based on metrics such as precision, recall, and $F_1$ score.

Expected Frequency Matrices of Elections: Computation, Geometry, and Preference Learning

We use the "map of elections" approach of Szufa et al. (AAMAS 2020) to analyze several well-known vote distributions. For each of them, we give an explicit formula or an efficient algorithm for computing its frequency matrix, which captures the probability that a given candidate appears in a given position in a sampled vote. We use these matrices to draw the "skeleton map" of distributions, evaluate its robustness, and analyze its properties. We further use them to identify the nature of several real-world elections.

FactPEGASUS: Factuality-Aware Pre-training and Fine-tuning for Abstractive Summarization

We present FactPEGASUS, an abstractive summarization model that addresses the problem of factuality during pre-training and fine-tuning: (1) We augment the sentence selection strategy of PEGASUS's (Zhang et al., 2020) pre-training objective to create pseudo-summaries that are both important and factual; (2) We introduce three complementary components for fine-tuning. The corrector removes hallucinations present in the reference summary, the contrastor uses contrastive learning to better differentiate nonfactual summaries from factual ones, and the connector bridges the gap between the pre-training and fine-tuning for better transfer of knowledge. Experiments on three downstream tasks demonstrate that FactPEGASUS substantially improves factuality evaluated by multiple automatic metrics and humans. Our thorough analysis suggests that FactPEGASUS is more factual than using the original pre-training objective in zero-shot and few-shot settings, retains factual behavior more robustly than strong baselines, and does not rely entirely on becoming more extractive to improve factuality. Our code and data are publicly available at: https://github.com/meetdavidwan/factpegasus

Federated Anomaly Detection over Distributed Data Streams

Sharing of telecommunication network data, for example, even at high aggregation levels, is nowadays highly restricted due to privacy legislation and regulations and other important ethical concerns. It leads to scattering data across institutions, regions, and states, inhibiting the usage of AI methods that could otherwise take advantage of data at scale. It creates the need to build a platform to control such data, build models or perform calculations. In this work, we propose an approach to building the bridge among anomaly detection, federated learning, and data streams. The overarching goal of the work is to detect anomalies in a federated environment over distributed data streams. This work complements the state-of-the-art by adapting the data stream algorithms in a federated learning setting for anomaly detection and by delivering a robust framework and demonstrating the practical feasibility in a real-world distributed deployment scenario.

GraphHD: Efficient graph classification using hyperdimensional computing

Hyperdimensional Computing (HDC) developed by Kanerva is a computational model for machine learning inspired by neuroscience. HDC exploits characteristics of biological neural systems such as high-dimensionality, randomness and a holographic representation of information to achieve a good balance between accuracy, efficiency and robustness. HDC models have already been proven to be useful in different learning applications, especially in resource-limited settings such as the increasingly popular Internet of Things (IoT). One class of learning tasks that is missing from the current body of work on HDC is graph classification. Graphs are among the most important forms of information representation, yet, to this day, HDC algorithms have not been applied to the graph learning problem in a general sense. Moreover, graph learning in IoT and sensor networks, with limited compute capabilities, introduce challenges to the overall design methodology. In this paper, we present GraphHD$-$a baseline approach for graph classification with HDC. We evaluate GraphHD on real-world graph classification problems. Our results show that when compared to the state-of-the-art Graph Neural Networks (GNNs) the proposed model achieves comparable accuracy, while training and inference times are on average 14.6$\times$ and 2.0$\times$ faster, respectively.

On the Difficulty of Defending Self-Supervised Learning against Model Extraction

Self-Supervised Learning (SSL) is an increasingly popular ML paradigm that trains models to transform complex inputs into representations without relying on explicit labels. These representations encode similarity structures that enable efficient learning of multiple downstream tasks. Recently, ML-as-a-Service providers have commenced offering trained SSL models over inference APIs, which transform user inputs into useful representations for a fee. However, the high cost involved to train these models and their exposure over APIs both make black-box extraction a realistic security threat. We thus explore model stealing attacks against SSL. Unlike traditional model extraction on classifiers that output labels, the victim models here output representations; these representations are of significantly higher dimensionality compared to the low-dimensional prediction scores output by classifiers. We construct several novel attacks and find that approaches that train directly on a victim's stolen representations are query efficient and enable high accuracy for downstream models. We then show that existing defenses against model extraction are inadequate and not easily retrofitted to the specificities of SSL.

CurFi: An automated tool to find the best regression analysis model using curve fitting

Regression analysis is a well known quantitative research method that primarily explores the relationship between one or more independent variables and a dependent variable. Conducting regression analysis manually on large datasets with multiple independent variables can be tedious. An automated system for regression analysis will be of great help for researchers as well as non-expert users. Thus, the objective of this research is to design and develop an automated curve fitting system. As outcome, a curve fitting system named "CurFi" was developed that uses linear regression models to fit a curve to a dataset and to find out the best fit model. The system facilitates to upload a dataset, split the dataset into training set and test set, select relevant features and label from the dataset; and the system will return the best fit linear regression model after training is completed. The developed tool would be a great resource for the users having limited technical knowledge who will also be able to find the best fit regression model for a dataset using the developed "CurFi" system.

The Primacy Bias in Deep Reinforcement Learning

This work identifies a common flaw of deep reinforcement learning (RL) algorithms: a tendency to rely on early interactions and ignore useful evidence encountered later. Because of training on progressively growing datasets, deep RL agents incur a risk of overfitting to earlier experiences, negatively affecting the rest of the learning process. Inspired by cognitive science, we refer to this effect as the primacy bias. Through a series of experiments, we dissect the algorithmic aspects of deep RL that exacerbate this bias. We then propose a simple yet generally-applicable mechanism that tackles the primacy bias by periodically resetting a part of the agent. We apply this mechanism to algorithms in both discrete (Atari 100k) and continuous action (DeepMind Control Suite) domains, consistently improving their performance.

Gradient-based Counterfactual Explanations using Tractable Probabilistic Models

Counterfactual examples are an appealing class of post-hoc explanations for machine learning models. Given input $x$ of class $y_1$, its counterfactual is a contrastive example $x^\prime$ of another class $y_0$. Current approaches primarily solve this task by a complex optimization: define an objective function based on the loss of the counterfactual outcome $y_0$ with hard or soft constraints, then optimize this function as a black-box. This "deep learning" approach, however, is rather slow, sometimes tricky, and may result in unrealistic counterfactual examples. In this work, we propose a novel approach to deal with these problems using only two gradient computations based on tractable probabilistic models. First, we compute an unconstrained counterfactual $u$ of $x$ to induce the counterfactual outcome $y_0$. Then, we adapt $u$ to higher density regions, resulting in $x^{\prime}$. Empirical evidence demonstrates the dominant advantages of our approach.

JR2net: A Joint Non-Linear Representation and Recovery Network for Compressive Spectral Imaging

Deep learning models are state-of-the-art in compressive spectral imaging (CSI) recovery. These methods use a deep neural network (DNN) as an image generator to learn non-linear mapping from compressed measurements to the spectral image. For instance, the deep spectral prior approach uses a convolutional autoencoder network (CAE) in the optimization algorithm to recover the spectral image by using a non-linear representation. However, the CAE training is detached from the recovery problem, which does not guarantee optimal representation of the spectral images for the CSI problem. This work proposes a joint non-linear representation and recovery network (JR2net), linking the representation and recovery task into a single optimization problem. JR2net consists of an optimization-inspired network following an ADMM formulation that learns a non-linear low-dimensional representation and simultaneously performs the spectral image recovery, trained via the end-to-end approach. Experimental results show the superiority of the proposed method with improvements up to 2.57 dB in PSNR and performance around 2000 times faster than state-of-the-art methods.

Efficient Algorithms for Planning with Participation Constraints

We consider the problem of planning with participation constraints introduced in [Zhang et al., 2022]. In this problem, a principal chooses actions in a Markov decision process, resulting in separate utilities for the principal and the agent. However, the agent can and will choose to end the process whenever his expected onward utility becomes negative. The principal seeks to compute and commit to a policy that maximizes her expected utility, under the constraint that the agent should always want to continue participating. We provide the first polynomial-time exact algorithm for this problem for finite-horizon settings, where previously only an additive $\varepsilon$-approximation algorithm was known. Our approach can also be extended to the (discounted) infinite-horizon case, for which we give an algorithm that runs in time polynomial in the size of the input and $\log(1/\varepsilon)$, and returns a policy that is optimal up to an additive error of $\varepsilon$.

Scalable algorithms for physics-informed neural and graph networks

Physics-informed machine learning (PIML) has emerged as a promising new approach for simulating complex physical and biological systems that are governed by complex multiscale processes for which some data are also available. In some instances, the objective is to discover part of the hidden physics from the available data, and PIML has been shown to be particularly effective for such problems for which conventional methods may fail. Unlike commercial machine learning where training of deep neural networks requires big data, in PIML big data are not available. Instead, we can train such networks from additional information obtained by employing the physical laws and evaluating them at random points in the space-time domain. Such physics-informed machine learning integrates multimodality and multifidelity data with mathematical models, and implements them using neural networks or graph networks. Here, we review some of the prevailing trends in embedding physics into machine learning, using physics-informed neural networks (PINNs) based primarily on feed-forward neural networks and automatic differentiation. For more complex systems or systems of systems and unstructured data, graph neural networks (GNNs) present some distinct advantages, and here we review how physics-informed learning can be accomplished with GNNs based on graph exterior calculus to construct differential operators; we refer to these architectures as physics-informed graph networks (PIGNs). We present representative examples for both forward and inverse problems and discuss what advances are needed to scale up PINNs, PIGNs and more broadly GNNs for large-scale engineering problems.

Data-Driven Interpolation for Super-Scarce X-Ray Computed Tomography

We address the problem of reconstructing X-Ray tomographic images from scarce measurements by interpolating missing acquisitions using a self-supervised approach. To do so, we train shallow neural networks to combine two neighbouring acquisitions into an estimated measurement at an intermediate angle. This procedure yields an enhanced sequence of measurements that can be reconstructed using standard methods, or further enhanced using regularisation approaches. Unlike methods that improve the sequence of acquisitions using an initial deterministic interpolation followed by machine-learning enhancement, we focus on inferring one measurement at once. This allows the method to scale to 3D, the computation to be faster and crucially, the interpolation to be significantly better than the current methods, when they exist. We also establish that a sequence of measurements must be processed as such, rather than as an image or a volume. We do so by comparing interpolation and up-sampling methods, and find that the latter significantly under-perform. We compare the performance of the proposed method against deterministic interpolation and up-sampling procedures and find that it outperforms them, even when used jointly with a state-of-the-art projection-data enhancement approach using machine-learning. These results are obtained for 2D and 3D imaging, on large biomedical datasets, in both projection space and image space.

On the inability of Gaussian process regression to optimally learn compositional functions

We rigorously prove that deep Gaussian process priors can outperform Gaussian process priors if the target function has a compositional structure. To this end, we study information-theoretic lower bounds for posterior contraction rates for Gaussian process regression in a continuous regression model. We show that if the true function is a generalized additive function, then the posterior based on any mean-zero Gaussian process can only recover the truth at a rate that is strictly slower than the minimax rate by a factor that is polynomially suboptimal in the sample size $n$.

Sharp Asymptotics of Self-training with Linear Classifier

Self-training (ST) is a straightforward and standard approach in semi-supervised learning, successfully applied to many machine learning problems. The performance of ST strongly depends on the supervised learning method used in the refinement step and the nature of the given data; hence, a general performance guarantee from a concise theory may become loose in a concrete setup. However, the theoretical methods that sharply predict how the performance of ST depends on various details for each learning scenario are limited. This study develops a novel theoretical framework for sharply characterizing the generalization abilities of the models trained by ST using the non-rigorous replica method of statistical physics. We consider the ST of the linear model that minimizes the ridge-regularized cross-entropy loss when the data are generated from a two-component Gaussian mixture. Consequently, we show that the generalization performance of ST in each iteration is sharply characterized by a small finite number of variables, which satisfy a set of deterministic self-consistent equations. By numerically solving these self-consistent equations, we find that ST's generalization performance approaches to the supervised learning method with a very simple regularization schedule when the label bias is small and a moderately large number of iterations are used.

Prioritizing Corners in OoD Detectors via Symbolic String Manipulation

For safety assurance of deep neural networks (DNNs), out-of-distribution (OoD) monitoring techniques are essential as they filter spurious input that is distant from the training dataset. This paper studies the problem of systematically testing OoD monitors to avoid cases where an input data point is tested as in-distribution by the monitor, but the DNN produces spurious output predictions. We consider the definition of "in-distribution" characterized in the feature space by a union of hyperrectangles learned from the training dataset. Thus the testing is reduced to finding corners in hyperrectangles distant from the available training data in the feature space. Concretely, we encode the abstract location of every data point as a finite-length binary string, and the union of all binary strings is stored compactly using binary decision diagrams (BDDs). We demonstrate how to use BDDs to symbolically extract corners distant from all data points within the training set. Apart from test case generation, we explain how to use the proposed corners to fine-tune the DNN to ensure that it does not predict overly confidently. The result is evaluated over examples such as number and traffic sign recognition.

An Artificial Neural Network Functionalized by Evolution

The topology of artificial neural networks has a significant effect on their performance. Characterizing efficient topology is a field of promising research in Artificial Intelligence. However, it is not a trivial task and it is mainly experimented on through convolutional neural networks. We propose a hybrid model which combines the tensor calculus of feed-forward neural networks with Pseudo-Darwinian mechanisms. This allows for finding topologies that are well adapted for elaboration of strategies, control problems or pattern recognition tasks. In particular, the model can provide adapted topologies at early evolutionary stages, and 'structural convergence', which can found applications in robotics, big-data and artificial life.

Pest presence prediction using interpretable machine learning

Helicoverpa Armigera, or cotton bollworm, is a serious insect pest of cotton crops that threatens the yield and the quality of lint. The timely knowledge of the presence of the insects in the field is crucial for effective farm interventions. Meteo-climatic and vegetation conditions have been identified as key drivers of crop pest abundance. In this work, we applied an interpretable classifier, i.e., Explainable Boosting Machine, which uses earth observation vegetation indices, numerical weather predictions and insect trap catches to predict the onset of bollworm harmfulness in cotton fields in Greece. The glass-box nature of our approach provides significant insight on the main drivers of the model and the interactions among them. Model interpretability adds to the trustworthiness of our approach and therefore its potential for rapid uptake and context-based implementation in operational farm management scenarios. Our results are satisfactory and the importance of drivers, through our analysis on global and local explainability, is in accordance with the literature.

Towards Space-to-Ground Data Availability for Agriculture Monitoring

The recent advances in machine learning and the availability of free and open big Earth data (e.g., Sentinel missions), which cover large areas with high spatial and temporal resolution, have enabled many agriculture monitoring applications. One example is the control of subsidy allocations of the Common Agricultural Policy (CAP). Advanced remote sensing systems have been developed towards the large-scale evidence-based monitoring of the CAP. Nevertheless, the spatial resolution of satellite images is not always adequate to make accurate decisions for all fields. In this work, we introduce the notion of space-to-ground data availability, i.e., from the satellite to the field, in an attempt to make the best out of the complementary characteristics of the different sources. We present a space-to-ground dataset that contains Sentinel-1 radar and Sentinel-2 optical image time-series, as well as street-level images from the crowdsourcing platform Mapillary, for grassland fields in the area of Utrecht for 2017. The multifaceted utility of our dataset is showcased through the downstream task of grassland classification. We train machine and deep learning algorithms on these different data domains and highlight the potential of fusion techniques towards increasing the reliability of decisions.

Generalizing to New Tasks via One-Shot Compositional Subgoals

The ability to generalize to previously unseen tasks with little to no supervision is a key challenge in modern machine learning research. It is also a cornerstone of a future "General AI". Any artificially intelligent agent deployed in a real world application, must adapt on the fly to unknown environments. Researchers often rely on reinforcement and imitation learning to provide online adaptation to new tasks, through trial and error learning. However, this can be challenging for complex tasks which require many timesteps or large numbers of subtasks to complete. These "long horizon" tasks suffer from sample inefficiency and can require extremely long training times before the agent can learn to perform the necessary longterm planning. In this work, we introduce CASE which attempts to address these issues by training an Imitation Learning agent using adaptive "near future" subgoals. These subgoals are recalculated at each step using compositional arithmetic in a learned latent representation space. In addition to improving learning efficiency for standard long-term tasks, this approach also makes it possible to perform one-shot generalization to previously unseen tasks, given only a single reference trajectory for the task in a different environment. Our experiments show that the proposed approach consistently outperforms the previous state-of-the-art compositional Imitation Learning approach by 30%.

Prediction of stent under-expansion in calcified coronary arteries using machine-learning on intravascular optical coherence tomography

BACKGROUND Careful evaluation of the risk of stent under-expansions before the intervention will aid treatment planning, including the application of a pre-stent plaque modification strategy. OBJECTIVES It remains challenging to achieve a proper stent expansion in the presence of severely calcified coronary lesions. Building on our work in deep learning segmentation, we created an automated machine learning approach that uses lesion attributes to predict stent under-expansion from pre-stent images, suggesting the need for plaque modification. METHODS Pre- and post-stent intravascular optical coherence tomography image data were obtained from 110 coronary lesions. Lumen and calcifications in pre-stent images were segmented using deep learning, and numerous features per lesion were extracted. We analyzed stent expansion along the lesion, enabling frame, segmental, and whole-lesion analyses. We trained regression models to predict the poststent lumen area and then to compute the stent expansion index (SEI). Stents with an SEI < or >/= 80% were classified as "under-expanded" and "well-expanded," respectively. RESULTS Best performance (root-mean-square-error = 0.04+/-0.02 mm2, r = 0.94+/-0.04, p < 0.0001) was achieved when we used features from both the lumen and calcification to train a Gaussian regression model for a segmental analysis over a segment length of 31 frames. Under-expansion classification results (AUC=0.85+/-0.02) were significantly improved over other approaches. CONCLUSIONS We used calcifications and lumen features to identify lesions at risk of stent under-expansion. Results suggest that the use of pre-stent images can inform physicians of the need to apply plaque modification approaches.

From Dirichlet to Rubin: Optimistic Exploration in RL without Bonuses

We propose the Bayes-UCBVI algorithm for reinforcement learning in tabular, stage-dependent, episodic Markov decision process: a natural extension of the Bayes-UCB algorithm by Kaufmann et al. (2012) for multi-armed bandits. Our method uses the quantile of a Q-value function posterior as upper confidence bound on the optimal Q-value function. For Bayes-UCBVI, we prove a regret bound of order $\widetilde{O}(\sqrt{H^3SAT})$ where $H$ is the length of one episode, $S$ is the number of states, $A$ the number of actions, $T$ the number of episodes, that matches the lower-bound of $\Omega(\sqrt{H^3SAT})$ up to poly-$\log$ terms in $H,S,A,T$ for a large enough $T$. To the best of our knowledge, this is the first algorithm that obtains an optimal dependence on the horizon $H$ (and $S$) without the need for an involved Bernstein-like bonus or noise. Crucial to our analysis is a new fine-grained anti-concentration bound for a weighted Dirichlet sum that can be of independent interest. We then explain how Bayes-UCBVI can be easily extended beyond the tabular setting, exhibiting a strong link between our algorithm and Bayesian bootstrap (Rubin, 1981).

Experimental Validation of Spectral-Spatial Power Evolution Design Using Raman Amplifiers

We experimentally validate a machine learning-enabled Raman amplification framework, capable of jointly shaping the signal power evolution in two domains: frequency and fiber distance. The proposed experiment addresses the amplification in the whole C-band, by optimizing four first-order counter-propagating Raman pumps.

Real-time semantic segmentation on FPGAs for autonomous vehicles with hls4ml

In this paper, we investigate how field programmable gate arrays can serve as hardware accelerators for real-time semantic segmentation tasks relevant for autonomous driving. Considering compressed versions of the ENet convolutional neural network architecture, we demonstrate a fully-on-chip deployment with a latency of 4.9 ms per image, using less than 30% of the available resources on a Xilinx ZCU102 evaluation board. The latency is reduced to 3 ms per image when increasing the batch size to ten, corresponding to the use case where the autonomous vehicle receives inputs from multiple cameras simultaneously. We show, through aggressive filter reduction and heterogeneous quantization-aware training, and an optimized implementation of convolutional layers, that the power consumption and resource utilization can be significantly reduced while maintaining accuracy on the Cityscapes dataset.

L3-Net Deep Audio Embeddings to Improve COVID-19 Detection from Smartphone Data

Smartphones and wearable devices, along with Artificial Intelligence, can represent a game-changer in the pandemic control, by implementing low-cost and pervasive solutions to recognize the development of new diseases at their early stages and by potentially avoiding the rise of new outbreaks. Some recent works show promise in detecting diagnostic signals of COVID-19 from voice and coughs by using machine learning and hand-crafted acoustic features. In this paper, we decided to investigate the capabilities of the recently proposed deep embedding model L3-Net to automatically extract meaningful features from raw respiratory audio recordings in order to improve the performances of standard machine learning classifiers in discriminating between COVID-19 positive and negative subjects from smartphone data. We evaluated the proposed model on 3 datasets, comparing the obtained results with those of two reference works. Results show that the combination of L3-Net with hand-crafted features overcomes the performance of the other works of 28.57% in terms of AUC in a set of subject-independent experiments. This result paves the way to further investigation on different deep audio embeddings, also for the automatic detection of different diseases.

Conditional Born machine for Monte Carlo events generation

Generative modeling is a promising task for near-term quantum devices, which can use the stochastic nature of quantum measurements as random source. So called Born machines are purely quantum models and promise to generate probability distributions in a quantum way, inaccessible to classical computers. This paper presents an application of Born machines to Monte Carlo simulations and extends their reach to multivariate and conditional distributions. Models are run on (noisy) simulators and IBM Quantum superconducting quantum hardware. More specifically, Born machines are used to generate muonic force carriers (MFC) events resulting from scattering processes between muons and the detector material in high-energy-physics colliders experiments. MFCs are bosons appearing in beyond the standard model theoretical frameworks, which are candidates for dark matter. Empirical evidences suggest that Born machines can reproduce the underlying distribution of datasets coming from Monte Carlo simulations, and are competitive with classical machine learning-based generative models of similar complexity.

Hyperdimensional computing encoding for feature selection on the use case of epileptic seizure detection

The healthcare landscape is moving from the reactive interventions focused on symptoms treatment to a more proactive prevention, from one-size-fits-all to personalized medicine, and from centralized to distributed paradigms. Wearable IoT devices and novel algorithms for continuous monitoring are essential components of this transition. Hyperdimensional (HD) computing is an emerging ML paradigm inspired by neuroscience research with various aspects interesting for IoT devices and biomedical applications. Here we explore the not yet addressed topic of optimal encoding of spatio-temporal data, such as electroencephalogram (EEG) signals, and all information it entails to the HD vectors. Further, we demonstrate how the HD computing framework can be used to perform feature selection by choosing an adequate encoding. To the best of our knowledge, this is the first approach to performing feature selection using HD computing in the literature. As a result, we believe it can support the ML community to further foster the research in multiple directions related to feature and channel selection, as well as model interpretability.

Generalizing to Evolving Domains with Latent Structure-Aware Sequential Autoencoder

Domain generalization aims to improve the generalization capability of machine learning systems to out-of-distribution (OOD) data. Existing domain generalization techniques embark upon stationary and discrete environments to tackle the generalization issue caused by OOD data. However, many real-world tasks in non-stationary environments (e.g. self-driven car system, sensor measures) involve more complex and continuously evolving domain drift, which raises new challenges for the problem of domain generalization. In this paper, we formulate the aforementioned setting as the problem of evolving domain generalization. Specifically, we propose to introduce a probabilistic framework called Latent Structure-aware Sequential Autoencoder (LSSAE) to tackle the problem of evolving domain generalization via exploring the underlying continuous structure in the latent space of deep neural networks, where we aim to identify two major factors namely covariate shift and concept shift accounting for distribution shift in non-stationary environments. Experimental results on both synthetic and real-world datasets show that LSSAE can lead to superior performances based on the evolving domain generalization setting.

Taming Continuous Posteriors for Latent Variational Dialogue Policies

Utilizing amortized variational inference for latent-action reinforcement learning (RL) has been shown to be an effective approach in Task-oriented Dialogue (ToD) systems for optimizing dialogue success. Until now, categorical posteriors have been argued to be one of the main drivers of performance. In this work we revisit Gaussian variational posteriors for latent-action RL and show that they can yield even better performance than categoricals. We achieve this by simplifying the training procedure and propose ways to regularize the latent dialogue policy to retain good response coherence. Using continuous latent representations our model achieves state of the art dialogue success rate on the MultiWOZ benchmark, and also compares well to categorical latent methods in response coherence.

Attacking and Defending Deep Reinforcement Learning Policies

Recent studies have shown that deep reinforcement learning (DRL) policies are vulnerable to adversarial attacks, which raise concerns about applications of DRL to safety-critical systems. In this work, we adopt a principled way and study the robustness of DRL policies to adversarial attacks from the perspective of robust optimization. Within the framework of robust optimization, optimal adversarial attacks are given by minimizing the expected return of the policy, and correspondingly a good defense mechanism should be realized by improving the worst-case performance of the policy. Considering that attackers generally have no access to the training environment, we propose a greedy attack algorithm, which tries to minimize the expected return of the policy without interacting with the environment, and a defense algorithm, which performs adversarial training in a max-min form. Experiments on Atari game environments show that our attack algorithm is more effective and leads to worse return of the policy than existing attack algorithms, and our defense algorithm yields policies more robust than existing defense methods to a range of adversarial attacks (including our proposed attack algorithm).

Model Agnostic Local Explanations of Reject

The application of machine learning based decision making systems in safety critical areas requires reliable high certainty predictions. Reject options are a common way of ensuring a sufficiently high certainty of predictions made by the system. While being able to reject uncertain samples is important, it is also of importance to be able to explain why a particular sample was rejected. However, explaining general reject options is still an open problem. We propose a model agnostic method for locally explaining arbitrary reject options by means of interpretable models and counterfactual explanations.

Rethinking Reinforcement Learning based Logic Synthesis

Recently, reinforcement learning has been used to address logic synthesis by formulating the operator sequence optimization problem as a Markov decision process. However, through extensive experiments, we find out that the learned policy makes decisions independent from the circuit features (i.e., states) and yields an operator sequence that is permutation invariant to some extent in terms of operators. Based on these findings, we develop a new RL-based method that can automatically recognize critical operators and generate common operator sequences generalizable to unseen circuits. Our algorithm is verified on both the EPFL benchmark, a private dataset and a circuit at industrial scale. Experimental results demonstrate that it achieves a good balance among delay, area and runtime, and is practical for industrial usage.

Qualitative Differences Between Evolutionary Strategies and Reinforcement Learning Methods for Control of Autonomous Agents

In this paper we analyze the qualitative differences between evolutionary strategies and reinforcement learning algorithms by focusing on two popular state-of-the-art algorithms: the OpenAI-ES evolutionary strategy and the Proximal Policy Optimization (PPO) reinforcement learning algorithm -- the most similar methods of the two families. We analyze how the methods differ with respect to: (i) general efficacy, (ii) ability to cope with sparse rewards, (iii) propensity/capacity to discover minimal solutions, (iv) dependency on reward shaping, and (v) ability to cope with variations of the environmental conditions. The analysis of the performance and of the behavioral strategies displayed by the agents trained with the two methods on benchmark problems enable us to demonstrate qualitative differences which were not identified in previous studies, to identify the relative weakness of the two methods, and to propose ways to ameliorate some of those weakness. We show that the characteristics of the reward function has a strong impact which vary qualitatively not only for the OpenAI-ES and the PPO but also for alternative reinforcement learning algorithms, thus demonstrating the importance of optimizing the characteristic of the reward function to the algorithm used.

Fundamental Laws of Binary Classification

Finding discriminant functions of minimum risk binary classification systems is a novel geometric locus problem -- that requires solving a system of fundamental locus equations of binary classification -- subject to deep-seated statistical laws. We show that a discriminant function of a minimum risk binary classification system is the solution of a locus equation that represents the geometric locus of the decision boundary of the system, wherein the discriminant function is connected to the decision boundary by an intrinsic eigen-coordinate system in such a manner that the discriminant function is represented by a geometric locus of a novel principal eigenaxis -- formed by a dual locus of likelihood components and principal eigenaxis components. We demonstrate that a minimum risk binary classification system acts to jointly minimize its eigenenergy and risk by locating a point of equilibrium wherein critical minimum eigenenergies exhibited by the system are symmetrically concentrated in such a manner that the geometric locus of the novel principal eigenaxis of the system exhibits symmetrical dimensions and densities, such that counteracting and opposing forces and influences of the system are symmetrically balanced with each other -- about the geometric center of the locus of the novel principal eigenaxis -- whereon the statistical fulcrum of the system is located. Thereby, a minimum risk binary classification system satisfies a state of statistical equilibrium wherein the total allowed eigenenergy and the expected risk exhibited by the system are jointly minimized within the decision space of the system, so that the system exhibits the minimum probability of classification error.

Chemical transformer compression for accelerating both training and inference of molecular modeling

Transformer models have been developed in molecular science with excellent performance in applications including quantitative structure-activity relationship (QSAR) and virtual screening (VS). Compared with other types of models, however, they are large, which results in a high hardware requirement to abridge time for both training and inference processes. In this work, cross-layer parameter sharing (CLPS), and knowledge distillation (KD) are used to reduce the sizes of transformers in molecular science. Both methods not only have competitive QSAR predictive performance as compared to the original BERT model, but also are more parameter efficient. Furthermore, by integrating CLPS and KD into a two-state chemical network, we introduce a new deep lite chemical transformer model, DeLiCaTe. DeLiCaTe captures general-domains as well as task-specific knowledge, which lead to a 4x faster rate of both training and inference due to a 10- and 3-times reduction of the number of parameters and layers, respectively. Meanwhile, it achieves comparable performance in QSAR and VS modeling. Moreover, we anticipate that the model compression strategy provides a pathway to the creation of effective generative transformer models for organic drug and material design.

Poincaré Heterogeneous Graph Neural Networks for Sequential Recommendation

Sequential recommendation (SR) learns users' preferences by capturing the sequential patterns from users' behaviors evolution. As discussed in many works, user-item interactions of SR generally present the intrinsic power-law distribution, which can be ascended to hierarchy-like structures. Previous methods usually handle such hierarchical information by making user-item sectionalization empirically under Euclidean space, which may cause distortion of user-item representation in real online scenarios. In this paper, we propose a Poincar\'{e}-based heterogeneous graph neural network named PHGR to model the sequential pattern information as well as hierarchical information contained in the data of SR scenarios simultaneously. Specifically, for the purpose of explicitly capturing the hierarchical information, we first construct a weighted user-item heterogeneous graph by aliening all the user-item interactions to improve the perception domain of each user from a global view. Then the output of the global representation would be used to complement the local directed item-item homogeneous graph convolution. By defining a novel hyperbolic inner product operator, the global and local graph representation learning are directly conducted in Poincar\'{e} ball instead of commonly used projection operation between Poincar\'{e} ball and Euclidean space, which could alleviate the cumulative error issue of general bidirectional translation process. Moreover, for the purpose of explicitly capturing the sequential dependency information, we design two types of temporal attention operations under Poincar\'{e} ball space. Empirical evaluations on datasets from the public and financial industry show that PHGR outperforms several comparison methods.

An Empirical Investigation of Representation Learning for Imitation

Imitation learning often needs a large demonstration set in order to handle the full range of situations that an agent might find itself in during deployment. However, collecting expert demonstrations can be expensive. Recent work in vision, reinforcement learning, and NLP has shown that auxiliary representation learning objectives can reduce the need for large amounts of expensive, task-specific data. Our Empirical Investigation of Representation Learning for Imitation (EIRLI) investigates whether similar benefits apply to imitation learning. We propose a modular framework for constructing representation learning algorithms, then use our framework to evaluate the utility of representation learning for imitation across several environment suites. In the settings we evaluate, we find that existing algorithms for image-based representation learning provide limited value relative to a well-tuned baseline with image augmentations. To explain this result, we investigate differences between imitation learning and other settings where representation learning has provided significant benefit, such as image classification. Finally, we release a well-documented codebase which both replicates our findings and provides a modular framework for creating new representation learning algorithms out of reusable components.

Weakly-supervised Biomechanically-constrained CT/MRI Registration of the Spine

CT and MRI are two of the most informative modalities in spinal diagnostics and treatment planning. CT is useful when analysing bony structures, while MRI gives information about the soft tissue. Thus, fusing the information of both modalities can be very beneficial. Registration is the first step for this fusion. While the soft tissues around the vertebra are deformable, each vertebral body is constrained to move rigidly. We propose a weakly-supervised deep learning framework that preserves the rigidity and the volume of each vertebra while maximizing the accuracy of the registration. To achieve this goal, we introduce anatomy-aware losses for training the network. We specifically design these losses to depend only on the CT label maps since automatic vertebra segmentation in CT gives more accurate results contrary to MRI. We evaluate our method on an in-house dataset of 167 patients. Our results show that adding the anatomy-aware losses increases the plausibility of the inferred transformation while keeping the accuracy untouched.

Autonomous Open-Ended Learning of Tasks with Non-Stationary Interdependencies

Autonomous open-ended learning is a relevant approach in machine learning and robotics, allowing the design of artificial agents able to acquire goals and motor skills without the necessity of user assigned tasks. A crucial issue for this approach is to develop strategies to ensure that agents can maximise their competence on as many tasks as possible in the shortest possible time. Intrinsic motivations have proven to generate a task-agnostic signal to properly allocate the training time amongst goals. While the majority of works in the field of intrinsically motivated open-ended learning focus on scenarios where goals are independent from each other, only few of them studied the autonomous acquisition of interdependent tasks, and even fewer tackled scenarios where goals involve non-stationary interdependencies. Building on previous works, we tackle these crucial issues at the level of decision making (i.e., building strategies to properly select between goals), and we propose a hierarchical architecture that treating sub-tasks selection as a Markov Decision Process is able to properly learn interdependent skills on the basis of intrinsically generated motivations. In particular, we first deepen the analysis of a previous system, showing the importance of incorporating information about the relationships between tasks at a higher level of the architecture (that of goal selection). Then we introduce H-GRAIL, a new system that extends the previous one by adding a new learning layer to store the autonomously acquired sequences of tasks to be able to modify them in case the interdependencies are non-stationary. All systems are tested in a real robotic scenario, with a Baxter robot performing multiple interdependent reaching tasks.

Towards on-sky adaptive optics control using reinforcement learning

The direct imaging of potentially habitable Exoplanets is one prime science case for the next generation of high contrast imaging instruments on ground-based extremely large telescopes. To reach this demanding science goal, the instruments are equipped with eXtreme Adaptive Optics (XAO) systems which will control thousands of actuators at a framerate of kilohertz to several kilohertz. Most of the habitable exoplanets are located at small angular separations from their host stars, where the current XAO systems' control laws leave strong residuals.Current AO control strategies like static matrix-based wavefront reconstruction and integrator control suffer from temporal delay error and are sensitive to mis-registration, i.e., to dynamic variations of the control system geometry. We aim to produce control methods that cope with these limitations, provide a significantly improved AO correction and, therefore, reduce the residual flux in the coronagraphic point spread function. We extend previous work in Reinforcement Learning for AO. The improved method, called PO4AO, learns a dynamics model and optimizes a control neural network, called a policy. We introduce the method and study it through numerical simulations of XAO with Pyramid wavefront sensing for the 8-m and 40-m telescope aperture cases. We further implemented PO4AO and carried out experiments in a laboratory environment using MagAO-X at the Steward laboratory. PO4AO provides the desired performance by improving the coronagraphic contrast in numerical simulations by factors 3-5 within the control region of DM and Pyramid WFS, in simulation and in the laboratory. The presented method is also quick to train, i.e., on timescales of typically 5-10 seconds, and the inference time is sufficiently small (< ms) to be used in real-time control for XAO with currently available hardware even for extremely large telescopes.

SQ-VAE: Variational Bayes on Discrete Representation with Self-annealed Stochastic Quantization

One noted issue of vector-quantized variational autoencoder (VQ-VAE) is that the learned discrete representation uses only a fraction of the full capacity of the codebook, also known as codebook collapse. We hypothesize that the training scheme of VQ-VAE, which involves some carefully designed heuristics, underlies this issue. In this paper, we propose a new training scheme that extends the standard VAE via novel stochastic dequantization and quantization, called stochastically quantized variational autoencoder (SQ-VAE). In SQ-VAE, we observe a trend that the quantization is stochastic at the initial stage of the training but gradually converges toward a deterministic quantization, which we call self-annealing. Our experiments show that SQ-VAE improves codebook utilization without using common heuristics. Furthermore, we empirically show that SQ-VAE is superior to VAE and VQ-VAE in vision- and speech-related tasks.

Automated Mobility Context Detection with Inertial Signals

Remote monitoring of motor functions is a powerful approach for health assessment, especially among the elderly population or among subjects affected by pathologies that negatively impact their walking capabilities. This is further supported by the continuous development of wearable sensor devices, which are getting progressively smaller, cheaper, and more energy efficient. The external environment and mobility context have an impact on walking performance, hence one of the biggest challenges when remotely analysing gait episodes is the ability to detect the context within which those episodes occurred. The primary goal of this paper is the investigation of context detection for remote monitoring of daily motor functions. We aim to understand whether inertial signals sampled with wearable accelerometers, provide reliable information to classify gait-related activities as either indoor or outdoor. We explore two different approaches to this task: (1) using gait descriptors and features extracted from the input inertial signals sampled during walking episodes, together with classic machine learning algorithms, and (2) treating the input inertial signals as time series data and leveraging end-to-end state-of-the-art time series classifiers. We directly compare the two approaches through a set of experiments based on data collected from 9 healthy individuals. Our results indicate that the indoor/outdoor context can be successfully derived from inertial data streams. We also observe that time series classification models achieve better accuracy than any other feature-based models, while preserving efficiency and ease of use.

Reachability Constrained Reinforcement Learning

Constrained reinforcement learning (CRL) has gained significant interest recently, since safety constraints satisfaction is critical for real-world problems. However, existing CRL methods constraining discounted cumulative costs generally lack rigorous definition and guarantee of safety. In contrast, in the safe control research, safety is defined as persistently satisfying certain state constraints. Such persistent safety is possible only on a subset of the state space, called feasible set, where an optimal largest feasible set exists for a given environment. Recent studies incorporate feasible sets into CRL with energy-based methods such as control barrier function (CBF), safety index (SI), and leverage prior conservative estimations of feasible sets, which harms the performance of the learned policy. To deal with this problem, this paper proposes the reachability CRL (RCRL) method by using reachability analysis to establish the novel self-consistency condition and characterize the feasible sets. The feasible sets are represented by the safety value function, which is used as the constraint in CRL. We use the multi-time scale stochastic approximation theory to prove that the proposed algorithm converges to a local optimum, where the largest feasible set can be guaranteed. Empirical results on different benchmarks validate the learned feasible set, the policy performance, and constraint satisfaction of RCRL, compared to CRL and safe control baselines.

Wasserstein t-SNE

Scientific datasets often have hierarchical structure: for example, in surveys, individual participants (samples) might be grouped at a higher level (units) such as their geographical region. In these settings, the interest is often in exploring the structure on the unit level rather than on the sample level. Units can be compared based on the distance between their means, however this ignores the within-unit distribution of samples. Here we develop an approach for exploratory analysis of hierarchical datasets using the Wasserstein distance metric that takes into account the shapes of within-unit distributions. We use t-SNE to construct 2D embeddings of the units, based on the matrix of pairwise Wasserstein distances between them. The distance matrix can be efficiently computed by approximating each unit with a Gaussian distribution, but we also provide a scalable method to compute exact Wasserstein distances. We use synthetic data to demonstrate the effectiveness of our Wasserstein t-SNE, and apply it to data from the 2017 German parliamentary election, considering polling stations as samples and voting districts as units. The resulting embedding uncovers meaningful structure in the data.

A model aggregation approach for high-dimensional large-scale optimization

Bayesian optimization (BO) has been widely used in machine learning and simulation optimization. With the increase in computational resources and storage capacities in these fields, high-dimensional and large-scale problems are becoming increasingly common. In this study, we propose a model aggregation method in the Bayesian optimization (MamBO) algorithm for efficiently solving high-dimensional large-scale optimization problems. MamBO uses a combination of subsampling and subspace embeddings to collectively address high dimensionality and large-scale issues; in addition, a model aggregation method is employed to address the surrogate model uncertainty issue that arises when embedding is applied. This surrogate model uncertainty issue is largely ignored in the embedding literature and practice, and it is exacerbated when the problem is high-dimensional and data are limited. Our proposed model aggregation method reduces these lower-dimensional surrogate model risks and improves the robustness of the BO algorithm. We derive an asymptotic bound for the proposed aggregated surrogate model and prove the convergence of MamBO. Benchmark numerical experiments indicate that our algorithm achieves superior or comparable performance to other commonly used high-dimensional BO algorithms. Moreover, we apply MamBO to a cascade classifier of a machine learning algorithm for face detection, and the results reveal that MamBO finds settings that achieve higher classification accuracy than the benchmark settings and is computationally faster than other high-dimensional BO algorithms.

A scalable deep learning approach for solving high-dimensional dynamic optimal transport

The dynamic formulation of optimal transport has attracted growing interests in scientific computing and machine learning, and its computation requires to solve a PDE-constrained optimization problem. The classical Eulerian discretization based approaches suffer from the curse of dimensionality, which arises from the approximation of high-dimensional velocity field. In this work, we propose a deep learning based method to solve the dynamic optimal transport in high dimensional space. Our method contains three main ingredients: a carefully designed representation of the velocity field, the discretization of the PDE constraint along the characteristics, and the computation of high dimensional integral by Monte Carlo method in each time step. Specifically, in the representation of the velocity field, we apply the classical nodal basis function in time and the deep neural networks in space domain with the H1-norm regularization. This technique promotes the regularity of the velocity field in both time and space such that the discretization along the characteristic remains to be stable during the training process. Extensive numerical examples have been conducted to test the proposed method. Compared to other solvers of optimal transport, our method could give more accurate results in high dimensional cases and has very good scalability with respect to dimension. Finally, we extend our method to more complicated cases such as crowd motion problem.

The use of deep learning in interventional radiotherapy (brachytherapy): a review with a focus on open source and open data

Deep learning advanced to one of the most important technologies in almost all medical fields. Especially in areas, related to medical imaging it plays a big role. However, in interventional radiotherapy (brachytherapy) deep learning is still in an early phase. In this review, first, we investigated and scrutinised the role of deep learning in all processes of interventional radiotherapy and directly related fields. Additionally we summarised the most recent developments. To reproduce results of deep learning algorithms both source code and training data must be available. Therefore, a second focus of this work was on the analysis of the availability of open source, open data and open models. In our analysis, we were able to show that deep learning plays already a major role in some areas of interventional radiotherapy, but is still hardly presented in others. Nevertheless, its impact is increasing with the years, partly self-propelled but also influenced by closely related fields. Open source, data and models are growing in number but are still scarce and unevenly distributed among different research groups. The reluctance in publishing code, data and models limits reproducibility and restricts evaluation to mono-institutional datasets. Summarised, deep learning will change positively the workflow of interventional radiotherapy but there is room for improvement when it comes to reproducible results and standardised evaluation methods.

KGRGRL: A User's Permission Reasoning Method Based on Knowledge Graph Reward Guidance Reinforcement Learning

In general, multiple domain cyberspace security assessments can be implemented by reasoning user's permissions. However, while existing methods include some information from the physical and social domains, they do not provide a comprehensive representation of cyberspace. Existing reasoning methods are also based on expert-given rules, resulting in inefficiency and a low degree of intelligence. To address this challenge, we create a Knowledge Graph (KG) of multiple domain cyberspace in order to provide a standard semantic description of the multiple domain cyberspace. Following that, we proposed a user's permissions reasoning method based on reinforcement learning. All permissions in cyberspace are represented as nodes, and an agent is trained to find all permissions that user can have according to user's initial permissions and cyberspace KG. We set 10 reward setting rules based on the features of cyberspace KG in the reinforcement learning of reward information setting, so that the agent can better locate user's all permissions and avoid blindly finding user's permissions. The results of the experiments showed that the proposed method can successfully reason about user's permissions and increase the intelligence level of the user's permissions reasoning method. At the same time, the F1 value of the proposed method is 6% greater than that of the Translating Embedding (TransE) method.

Multi-scale Attention Flow for Probabilistic Time Series Forecasting

The probability prediction of multivariate time series is a notoriously challenging but practical task. On the one hand, the challenge is how to effectively capture the cross-series correlations between interacting time series, to achieve accurate distribution modeling. On the other hand, we should consider how to capture the contextual information within time series more accurately to model multivariate temporal dynamics of time series. In this work, we proposed a novel non-autoregressive deep learning model, called Multi-scale Attention Normalizing Flow(MANF), where we integrate multi-scale attention and relative position information and the multivariate data distribution is represented by the conditioned normalizing flow. Additionally, compared with autoregressive modeling methods, our model avoids the influence of cumulative error and does not increase the time complexity. Extensive experiments demonstrate that our model achieves state-of-the-art performance on many popular multivariate datasets.

Robust Testing in High-Dimensional Sparse Models

We consider the problem of robustly testing the norm of a high-dimensional sparse signal vector under two different observation models. In the first model, we are given $n$ i.i.d. samples from the distribution $\mathcal{N}\left(\theta,I_d\right)$ (with unknown $\theta$), of which a small fraction has been arbitrarily corrupted. Under the promise that $\|\theta\|_0\le s$, we want to correctly distinguish whether $\|\theta\|_2=0$ or $\|\theta\|_2>\gamma$, for some input parameter $\gamma>0$. We show that any algorithm for this task requires $n=\Omega\left(s\log\frac{ed}{s}\right)$ samples, which is tight up to logarithmic factors. We also extend our results to other common notions of sparsity, namely, $\|\theta\|_q\le s$ for any $0 < q < 2$. In the second observation model that we consider, the data is generated according to a sparse linear regression model, where the covariates are i.i.d. Gaussian and the regression coefficient (signal) is known to be $s$-sparse. Here too we assume that an $\epsilon$-fraction of the data is arbitrarily corrupted. We show that any algorithm that reliably tests the norm of the regression coefficient requires at least $n=\Omega\left(\min(s\log d,{1}/{\gamma^4})\right)$ samples. Our results show that the complexity of testing in these two settings significantly increases under robustness constraints. This is in line with the recent observations made in robust mean testing and robust covariance testing.

Multiscale reconstruction of porous media based on multiple dictionaries learning

Digital modeling of the microstructure is important for studying the physical and transport properties of porous media. Multiscale modeling for porous media can accurately characterize macro-pores and micro-pores in a large-FoV (field of view) high-resolution three-dimensional pore structure model. This paper proposes a multiscale reconstruction algorithm based on multiple dictionaries learning, in which edge patterns and micro-pore patterns from homology high-resolution pore structure are introduced into low-resolution pore structure to build a fine multiscale pore structure model. The qualitative and quantitative comparisons of the experimental results show that the results of multiscale reconstruction are similar to the real high-resolution pore structure in terms of complex pore geometry and pore surface morphology. The geometric, topological and permeability properties of multiscale reconstruction results are almost identical to those of the real high-resolution pore structures. The experiments also demonstrate the proposal algorithm is capable of multiscale reconstruction without regard to the size of the input. This work provides an effective method for fine multiscale modeling of porous media.

Learning-Based sensitivity analysis and feedback design for drug delivery of mixed therapy of cancer in the presence of high model uncertainties

In this paper, a methodology is proposed that enables to analyze the sensitivity of the outcome of a therapy to unavoidable high dispersion of the patient specific parameters on one hand and to the choice of the parameters that define the drug delivery feedback strategy on the other hand. More precisely, a method is given that enables to extract and rank the most influent parameters that determine the probability of success/failure of a given feedback therapy for a given set of initial conditions over a cloud of realizations of uncertainties. Moreover predictors of the expectations of the amounts of drugs being used can also be derived. This enables to design an efficient stochastic optimization framework that guarantees safe contraction of the tumor while minimizing a weighted sum of the quantities of the different drugs being used. The framework is illustrated and validated using the example of a mixed therapy of cancer involving three combined drugs namely: a chemotherapy drug, an immunology vaccine and an immunotherapy drug. Finally, in this specific case, it is shown that dash-boards can be built in the 2D-space of the most influent state components that summarize the outcomes' probabilities and the associated drug usage as iso-values curves in the reduced state space.

Manifold Characteristics That Predict Downstream Task Performance

Pretraining methods are typically compared by evaluating the accuracy of linear classifiers, transfer learning performance, or visually inspecting the representation manifold's (RM) lower-dimensional projections. We show that the differences between methods can be understood more clearly by investigating the RM directly, which allows for a more detailed comparison. To this end, we propose a framework and new metric to measure and compare different RMs. We also investigate and report on the RM characteristics for various pretraining methods. These characteristics are measured by applying sequentially larger local alterations to the input data, using white noise injections and Projected Gradient Descent (PGD) adversarial attacks, and then tracking each datapoint. We calculate the total distance moved for each datapoint and the relative change in distance between successive alterations. We show that self-supervised methods learn an RM where alterations lead to large but constant size changes, indicating a smoother RM than fully supervised methods. We then combine these measurements into one metric, the Representation Manifold Quality Metric (RMQM), where larger values indicate larger and less variable step sizes, and show that RMQM correlates positively with performance on downstream tasks.

Ergodic variational flows

This work presents a new class of variational family -- ergodic variational flows -- that not only enables tractable i.i.d. sampling and density evaluation, but also comes with MCMC-like convergence guarantees. Ergodic variational flows consist of a mixture of repeated applications of a measure-preserving and ergodic map to an initial reference distribution. We provide mild conditions under which the variational distribution converges weakly and in total variation to the target as the number of steps in the flow increases; this convergence holds regardless of the value of variational parameters, although different parameter values may result in faster or slower convergence. Further, we develop a particular instantiation of the general family using Hamiltonian dynamics combined with deterministic momentum refreshment. Simulated and real data experiments provide an empirical verification of the convergence theory and demonstrate that samples produced by the method are of comparable quality to a state-of-the-art MCMC method.

Towards Lossless ANN-SNN Conversion under Ultra-Low Latency with Dual-Phase Optimization

Spiking neural network (SNN) operating with asynchronous discrete events shows higher energy efficiency. A popular approach to implement deep SNNs is ANN-SNN conversion combining both efficient training in ANNs and efficient inference in SNNs. However, the previous works mostly required thousands of time steps to achieve lossless conversion. In this paper, we first identify the underlying cause, i.e., misrepresentation of the negative or overflow residual membrane potential in SNNs. Furthermore, we systematically analyze the conversion error between SNNs and ANNs, and then decompose it into three folds: quantization error, clipping error, and residual membrane potential representation error. With such insights, we propose a dual-phase conversion algorithm to minimize those errors. As a result, our model achieves SOTA in both accuracy and accuracy-delay tradeoff with deep architectures (ResNet and VGG net). Specifically, we report SOTA accuracy within 16$\times$ speedup compared with the latest results. Meanwhile, lossless conversion is performed with at least 2$\times$ faster reasoning performance.

$q$-Munchausen Reinforcement Learning

The recently successful Munchausen Reinforcement Learning (M-RL) features implicit Kullback-Leibler (KL) regularization by augmenting the reward function with logarithm of the current stochastic policy. Though significant improvement has been shown with the Boltzmann softmax policy, when the Tsallis sparsemax policy is considered, the augmentation leads to a flat learning curve for almost every problem considered. We show that it is due to the mismatch between the conventional logarithm and the non-logarithmic (generalized) nature of Tsallis entropy. Drawing inspiration from the Tsallis statistics literature, we propose to correct the mismatch of M-RL with the help of $q$-logarithm/exponential functions. The proposed formulation leads to implicit Tsallis KL regularization under the maximum Tsallis entropy framework. We show such formulation of M-RL again achieves superior performance on benchmark problems and sheds light on more general M-RL with various entropic indices $q$.

Gradient Descent Optimizes Infinite-Depth ReLU Implicit Networks with Linear Widths

Implicit deep learning has recently become popular in the machine learning community since these implicit models can achieve competitive performance with state-of-the-art deep networks while using significantly less memory and computational resources. However, our theoretical understanding of when and how first-order methods such as gradient descent (GD) converge on \textit{nonlinear} implicit networks is limited. Although this type of problem has been studied in standard feed-forward networks, the case of implicit models is still intriguing because implicit networks have \textit{infinitely} many layers. The corresponding equilibrium equation probably admits no or multiple solutions during training. This paper studies the convergence of both gradient flow (GF) and gradient descent for nonlinear ReLU activated implicit networks. To deal with the well-posedness problem, we introduce a fixed scalar to scale the weight matrix of the implicit layer and show that there exists a small enough scaling constant, keeping the equilibrium equation well-posed throughout training. As a result, we prove that both GF and GD converge to a global minimum at a linear rate if the width $m$ of the implicit network is \textit{linear} in the sample size $N$, i.e., $m=\Omega(N)$.

Diffusion Models for Adversarial Purification

Adversarial purification refers to a class of defense methods that remove adversarial perturbations using a generative model. These methods do not make assumptions on the form of attack and the classification model, and thus can defend pre-existing classifiers against unseen threats. However, their performance currently falls behind adversarial training methods. In this work, we propose DiffPure that uses diffusion models for adversarial purification: Given an adversarial example, we first diffuse it with a small amount of noise following a forward diffusion process, and then recover the clean image through a reverse generative process. To evaluate our method against strong adaptive attacks in an efficient and scalable way, we propose to use the adjoint method to compute full gradients of the reverse generative process. Extensive experiments on three image datasets including CIFAR-10, ImageNet and CelebA-HQ with three classifier architectures including ResNet, WideResNet and ViT demonstrate that our method achieves the state-of-the-art results, outperforming current adversarial training and adversarial purification methods, often by a large margin. Project page: https://diffpure.github.io.

Miutsu: NTU's TaskBot for the Alexa Prize

This paper introduces Miutsu, National Taiwan University's Alexa Prize TaskBot, which is designed to assist users in completing tasks requiring multiple steps and decisions in two different domains -- home improvement and cooking. We overview our system design and architectural goals, and detail the proposed core elements, including question answering, task retrieval, social chatting, and various conversational modules. A dialogue flow is proposed to provide a robust and engaging conversation when handling complex tasks. We discuss the faced challenges during the competition and potential future work.

A Deep Reinforcement Learning Blind AI in DareFightingICE

This paper presents a deep reinforcement learning AI that uses sound as the input on the DareFightingICE platform at the DareFightingICE Competition in IEEE CoG 2022. In this work, an AI that only uses sound as the input is called blind AI. While state-of-the-art AIs rely mostly on visual or structured observations provided by their environments, learning to play games from only sound is still new and thus challenging. We propose different approaches to process audio data and use the Proximal Policy Optimization algorithm for our blind AI. We also propose to use our blind AI in evaluation of sound designs submitted to the competition and define three metrics for this task. The experimental results show the effectiveness of not only our blind AI but also the proposed three metrics.

Enforcing KL Regularization in General Tsallis Entropy Reinforcement Learning via Advantage Learning

Maximum Tsallis entropy (MTE) framework in reinforcement learning has gained popularity recently by virtue of its flexible modeling choices including the widely used Shannon entropy and sparse entropy. However, non-Shannon entropies suffer from approximation error and subsequent underperformance either due to its sensitivity or the lack of closed-form policy expression. To improve the tradeoff between flexibility and empirical performance, we propose to strengthen their error-robustness by enforcing implicit Kullback-Leibler (KL) regularization in MTE motivated by Munchausen DQN (MDQN). We do so by drawing connection between MDQN and advantage learning, by which MDQN is shown to fail on generalizing to the MTE framework. The proposed method Tsallis Advantage Learning (TAL) is verified on extensive experiments to not only significantly improve upon Tsallis-DQN for various non-closed-form Tsallis entropies, but also exhibits comparable performance to state-of-the-art maximum Shannon entropy algorithms.

Optimizing the optimizer for data driven deep neural networks and physics informed neural networks

We investigate the role of the optimizer in determining the quality of the model fit for neural networks with a small to medium number of parameters. We study the performance of Adam, an algorithm for first-order gradient-based optimization that uses adaptive momentum, the Levenberg and Marquardt (LM) algorithm a second order method, Broyden,Fletcher,Goldfarb and Shanno algorithm (BFGS) a second order method and LBFGS, a low memory version of BFGS. Using these optimizers we fit the function y = sinc(10x) using a neural network with a few parameters. This function has a variable amplitude and a constant frequency. We observe that the higher amplitude components of the function are fitted first and the Adam, BFGS and LBFGS struggle to fit the lower amplitude components of the function. We also solve the Burgers equation using a physics informed neural network(PINN) with the BFGS and LM optimizers. For our example problems with a small to medium number of weights, we find that the LM algorithm is able to rapidly converge to machine precision offering significant benefits over other optimizers. We further investigated the Adam optimizer with a range of models and found that Adam optimiser requires much deeper models with large numbers of hidden units containing up to 26x more parameters, in order to achieve a model fit close that achieved by the LM optimizer. The LM optimizer results illustrate that it may be possible build models with far fewer parameters. We have implemented all our methods in Keras and TensorFlow 2.

Explanation-Guided Fairness Testing through Genetic Algorithm

The fairness characteristic is a critical attribute of trusted AI systems. A plethora of research has proposed diverse methods for individual fairness testing. However, they are suffering from three major limitations, i.e., low efficiency, low effectiveness, and model-specificity. This work proposes ExpGA, an explanationguided fairness testing approach through a genetic algorithm (GA). ExpGA employs the explanation results generated by interpretable methods to collect high-quality initial seeds, which are prone to derive discriminatory samples by slightly modifying feature values. ExpGA then adopts GA to search discriminatory sample candidates by optimizing a fitness value. Benefiting from this combination of explanation results and GA, ExpGA is both efficient and effective to detect discriminatory individuals. Moreover, ExpGA only requires prediction probabilities of the tested model, resulting in a better generalization capability to various models. Experiments on multiple real-world benchmarks, including tabular and text datasets, show that ExpGA presents higher efficiency and effectiveness than four state-of-the-art approaches.

On the Convergence of the Shapley Value in Parametric Bayesian Learning Games

Measuring contributions is a classical problem in cooperative game theory where the Shapley value is the most well-known solution concept. In this paper, we establish the convergence property of the Shapley value in parametric Bayesian learning games where players perform a Bayesian inference using their combined data, and the posterior-prior KL divergence is used as the characteristic function. We show that for any two players, under some regularity conditions, their difference in Shapley value converges in probability to the difference in Shapley value of a limiting game whose characteristic function is proportional to the log-determinant of the joint Fisher information. As an application, we present an online collaborative learning framework that is asymptotically Shapley-fair. Our result enables this to be achieved without any costly computations of posterior-prior KL divergences. Only a consistent estimator of the Fisher information is needed. The effectiveness of our framework is demonstrated with experiments using real-world data.

Exploring the Learning Difficulty of Data Theory and Measure

As learning difficulty is crucial for machine learning (e.g., difficulty-based weighting learning strategies), previous literature has proposed a number of learning difficulty measures. However, no comprehensive investigation for learning difficulty is available to date, resulting in that nearly all existing measures are heuristically defined without a rigorous theoretical foundation. In addition, there is no formal definition of easy and hard samples even though they are crucial in many studies. This study attempts to conduct a pilot theoretical study for learning difficulty of samples. First, a theoretical definition of learning difficulty is proposed on the basis of the bias-variance trade-off theory on generalization error. Theoretical definitions of easy and hard samples are established on the basis of the proposed definition. A practical measure of learning difficulty is given as well inspired by the formal definition. Second, the properties for learning difficulty-based weighting strategies are explored. Subsequently, several classical weighting methods in machine learning can be well explained on account of explored properties. Third, the proposed measure is evaluated to verify its reasonability and superiority in terms of several main difficulty factors. The comparison in these experiments indicates that the proposed measure significantly outperforms the other measures throughout the experiments.

Optimal Randomized Approximations for Matrix based Renyi's Entropy

The Matrix-based Renyi's entropy enables us to directly measure information quantities from given data without the costly probability density estimation of underlying distributions, thus has been widely adopted in numerous statistical learning and inference tasks. However, exactly calculating this new information quantity requires access to the eigenspectrum of a semi-positive definite (SPD) matrix $A$ which grows linearly with the number of samples $n$, resulting in a $O(n^3)$ time complexity that is prohibitive for large-scale applications. To address this issue, this paper takes advantage of stochastic trace approximations for matrix-based Renyi's entropy with arbitrary $\alpha \in R^+$ orders, lowering the complexity by converting the entropy approximation to a matrix-vector multiplication problem. Specifically, we develop random approximations for integer order $\alpha$ cases and polynomial series approximations (Taylor and Chebyshev) for non-integer $\alpha$ cases, leading to a $O(n^2sm)$ overall time complexity, where $s,m \ll n$ denote the number of vector queries and the polynomial order respectively. We theoretically establish statistical guarantees for all approximation algorithms and give explicit order of s and m with respect to the approximation error $\varepsilon$, showing optimal convergence rate for both parameters up to a logarithmic factor. Large-scale simulations and real-world applications validate the effectiveness of the developed approximations, demonstrating remarkable speedup with negligible loss in accuracy.

Trustworthy Graph Neural Networks: Aspects, Methods and Trends

Graph neural networks (GNNs) have emerged as a series of competent graph learning methods for diverse real-world scenarios, ranging from daily applications like recommendation systems and question answering to cutting-edge technologies such as drug discovery in life sciences and n-body simulation in astrophysics. However, task performance is not the only requirement for GNNs. Performance-oriented GNNs have exhibited potential adverse effects like vulnerability to adversarial attacks, unexplainable discrimination against disadvantaged groups, or excessive resource consumption in edge computing environments. To avoid these unintentional harms, it is necessary to build competent GNNs characterised by trustworthiness. To this end, we propose a comprehensive roadmap to build trustworthy GNNs from the view of the various computing technologies involved. In this survey, we introduce basic concepts and comprehensively summarise existing efforts for trustworthy GNNs from six aspects, including robustness, explainability, privacy, fairness, accountability, and environmental well-being. Additionally, we highlight the intricate cross-aspect relations between the above six aspects of trustworthy GNNs. Finally, we present a thorough overview of trending directions for facilitating the research and industrialisation of trustworthy GNNs.

TNN7: A Custom Macro Suite for Implementing Highly Optimized Designs of Neuromorphic TNNs

Temporal Neural Networks (TNNs), inspired from the mammalian neocortex, exhibit energy-efficient online sensory processing capabilities. Recent works have proposed a microarchitecture framework for implementing TNNs and demonstrated competitive performance on vision and time-series applications. Building on these previous works, this work proposes TNN7, a suite of nine highly optimized custom macros developed using a predictive 7nm Process Design Kit (PDK), to enhance the efficiency, modularity and flexibility of the TNN design framework. TNN prototypes for two applications are used for evaluation of TNN7. An unsupervised time-series clustering TNN delivering competitive performance can be implemented within 40 uW power and 0.05 mm^2 area, while a 4-layer TNN that achieves an MNIST error rate of 1% consumes only 18 mW and 24.63 mm^2. On average, the proposed macros reduce power, delay, area, and energy-delay product by 14%, 16%, 28%, and 45%, respectively. Furthermore, employing TNN7 significantly reduces the synthesis runtime of TNN designs (by more than 3x), allowing for highly-scaled TNN implementations to be realized.

Training neural networks using Metropolis Monte Carlo and an adaptive variant

We examine the zero-temperature Metropolis Monte Carlo algorithm as a tool for training a neural network by minimizing a loss function. We find that, as expected on theoretical grounds and shown empirically by other authors, Metropolis Monte Carlo can train a neural net with an accuracy comparable to that of gradient descent, if not necessarily as quickly. The Metropolis algorithm does not fail automatically when the number of parameters of a neural network is large. It can fail when a neural network's structure or neuron activations are strongly heterogenous, and we introduce an adaptive Monte Carlo algorithm, aMC, to overcome these limitations. The intrinsic stochasticity of the Monte Carlo method allows aMC to train neural networks in which the gradient is too small to allow training by gradient descent. We suggest that, as for molecular simulation, Monte Carlo methods offer a complement to gradient-based methods for training neural networks, allowing access to a distinct set of network architectures and principles.

What GPT Knows About Who is Who

Coreference resolution -- which is a crucial task for understanding discourse and language at large -- has yet to witness widespread benefits from large language models (LLMs). Moreover, coreference resolution systems largely rely on supervised labels, which are highly expensive and difficult to annotate, thus making it ripe for prompt engineering. In this paper, we introduce a QA-based prompt-engineering method and discern \textit{generative}, pre-trained LLMs' abilities and limitations toward the task of coreference resolution. Our experiments show that GPT-2 and GPT-Neo can return valid answers, but that their capabilities to identify coreferent mentions are limited and prompt-sensitive, leading to inconsistent results.

SuperWarp: Supervised Learning and Warping on U-Net for Invariant Subvoxel-Precise Registration

In recent years, learning-based image registration methods have gradually moved away from direct supervision with target warps to instead use self-supervision, with excellent results in several registration benchmarks. These approaches utilize a loss function that penalizes the intensity differences between the fixed and moving images, along with a suitable regularizer on the deformation. In this paper, we argue that the relative failure of supervised registration approaches can in part be blamed on the use of regular U-Nets, which are jointly tasked with feature extraction, feature matching, and estimation of deformation. We introduce one simple but crucial modification to the U-Net that disentangles feature extraction and matching from deformation prediction, allowing the U-Net to warp the features, across levels, as the deformation field is evolved. With this modification, direct supervision using target warps begins to outperform self-supervision approaches that require segmentations, presenting new directions for registration when images do not have segmentations. We hope that our findings in this preliminary workshop paper will re-ignite research interest in supervised image registration techniques. Our code is publicly available from https://github.com/balbasty/superwarp.

Sibyl: Adaptive and Extensible Data Placement in Hybrid Storage Systems Using Online Reinforcement Learning

Hybrid storage systems (HSS) use multiple different storage devices to provide high and scalable storage capacity at high performance. Recent research proposes various techniques that aim to accurately identify performance-critical data to place it in a "best-fit" storage device. Unfortunately, most of these techniques are rigid, which (1) limits their adaptivity to perform well for a wide range of workloads and storage device configurations, and (2) makes it difficult for designers to extend these techniques to different storage system configurations (e.g., with a different number or different types of storage devices) than the configuration they are designed for. We introduce Sibyl, the first technique that uses reinforcement learning for data placement in hybrid storage systems. Sibyl observes different features of the running workload as well as the storage devices to make system-aware data placement decisions. For every decision it makes, Sibyl receives a reward from the system that it uses to evaluate the long-term performance impact of its decision and continuously optimizes its data placement policy online. We implement Sibyl on real systems with various HSS configurations. Our results show that Sibyl provides 21.6%/19.9% performance improvement in a performance-oriented/cost-oriented HSS configuration compared to the best previous data placement technique. Our evaluation using an HSS configuration with three different storage devices shows that Sibyl outperforms the state-of-the-art data placement policy by 23.9%-48.2%, while significantly reducing the system architect's burden in designing a data placement mechanism that can simultaneously incorporate three storage devices. We show that Sibyl achieves 80% of the performance of an oracle policy that has complete knowledge of future access patterns while incurring a very modest storage overhead of only 124.4 KiB.

Inverse design of nano-photonic wavelength demultiplexer with a deep neural network approach

In this paper, we propose a pre-trained-combined neural network (PTCN) as a comprehensive solution to the inverse design of an integrated photonic circuit. By utilizing both the initially pre-trained inverse and forward model with a joint training process, our PTCN model shows remarkable tolerance to the quantity and quality of the training data. As a proof of concept demonstration, the inverse design of a wavelength demultiplexer is used to verify the effectiveness of the PTCN model. The correlation coefficient of the prediction by the presented PTCN model remains greater than 0.974 even when the size of training data is decreased to 17%. The experimental results show a good agreement with predictions, and demonstrate a wavelength demultiplexer with an ultra-compact footprint, a high transmission efficiency with a transmission loss of -2dB, a low reflection of -10dB, and low crosstalk around -7dB simultaneously.

Learning Representations for New Sound Classes With Continual Self-Supervised Learning

In this paper, we present a self-supervised learning framework for continually learning representations for new sound classes. The proposed system relies on a continually trained neural encoder that is trained with similarity-based learning objectives without using labels. We show that representations learned with the proposed method generalize better and are less susceptible to catastrophic forgetting than fully-supervised approaches. Remarkably, our technique does not store past data or models and is more computationally efficient than distillation-based methods. To accurately assess the system performance, in addition to using existing protocols, we propose two realistic evaluation protocols that use only a small amount of labeled data to simulate practical use cases.

Incorporating Prior Knowledge into Neural Networks through an Implicit Composite Kernel

It is challenging to guide neural network (NN) learning with prior knowledge. In contrast, many known properties, such as spatial smoothness or seasonality, are straightforward to model by choosing an appropriate kernel in a Gaussian process (GP). Many deep learning applications could be enhanced by modeling such known properties. For example, convolutional neural networks (CNNs) are frequently used in remote sensing, which is subject to strong seasonal effects. We propose to blend the strengths of deep learning and the clear modeling capabilities of GPs by using a composite kernel that combines a kernel implicitly defined by a neural network with a second kernel function chosen to model known properties (e.g., seasonality). Then, we approximate the resultant GP by combining a deep network and an efficient mapping based on the Nystrom approximation, which we call Implicit Composite Kernel (ICK). ICK is flexible and can be used to include prior information in neural networks in many applications. We demonstrate the strength of our framework by showing its superior performance and flexibility on both synthetic and real-world data sets. The code is available at: https://anonymous.4open.science/r/ICK_NNGP-17C5/.

The Splendors and Miseries of Heavisidisation

Machine Learning (ML) is applicable to scientific problems, i.e. to those which have a well defined answer, only if this answer can be brought to a peculiar form ${\cal G}: X\longrightarrow Z$ with ${\cal G}(\vec x)$ expressed as a combination of iterated Heaviside functions. At present it is far from obvious, if and when such representations exist, what are the obstacles and, if they are absent, what are the ways to convert the known formulas into this form. This gives rise to a program of reformulation of ordinary science in such terms -- which sounds like a strong enhancement of the constructive mathematics approach, only this time it concerns all natural sciences. We describe the first steps on this long way.

Effect of Batch Normalization on Noise Resistant Property of Deep Learning Models

The fast execution speed and energy efficiency of analog hardware has made them a strong contender for deployment of deep learning model at the edge. However, there are concerns about the presence of analog noise which causes changes to the weight of the models, leading to performance degradation of deep learning model, despite their inherent noise resistant characteristics. The effect of the popular batch normalization layer on the noise resistant ability of deep learning model is investigated in this work. This systematic study has been carried out by first training different models with and without batch normalization layer on CIFAR10 and CIFAR100 dataset. The weights of the resulting models are then injected with analog noise and the performance of the models on the test dataset is obtained and compared. The results show that the presence of batch normalization layer negatively impacts noise resistant property of deep learning model and the impact grows with the increase of the number of batch normalization layers.

High-Resolution CMB Lensing Reconstruction with Deep Learning

Next-generation cosmic microwave background (CMB) surveys are expected to provide valuable information about the primordial universe by creating maps of the mass along the line of sight. Traditional tools for creating these lensing convergence maps include the quadratic estimator and the maximum likelihood based iterative estimator. Here, we apply a generative adversarial network (GAN) to reconstruct the lensing convergence field. We compare our results with a previous deep learning approach -- Residual-UNet -- and discuss the pros and cons of each. In the process, we use training sets generated by a variety of power spectra, rather than the one used in testing the methods.

What is an equivariant neural network?

We explain equivariant neural networks, a notion underlying breakthroughs in machine learning from deep convolutional neural networks for computer vision to AlphaFold 2 for protein structure prediction, without assuming knowledge of equivariance or neural networks. The basic mathematical ideas are simple but are often obscured by engineering complications that come with practical realizations. We extract and focus on the mathematical aspects, and limit ourselves to a cursory treatment of the engineering issues at the end.

Novel Multicolumn Kernel Extreme Learning Machine for Food Detection via Optimal Features from CNN

Automatic food detection is an emerging topic of interest due to its wide array of applications ranging from detecting food images on social media platforms to filtering non-food photos from the users in dietary assessment apps. Recently, during the COVID-19 pandemic, it has facilitated enforcing an eating ban by automatically detecting eating activities from cameras in public places. Therefore, to tackle the challenge of recognizing food images with high accuracy, we proposed the idea of a hybrid framework for extracting and selecting optimal features from an efficient neural network. There on, a nonlinear classifier is employed to discriminate between linearly inseparable feature vectors with great precision. In line with this idea, our method extracts features from MobileNetV3, selects an optimal subset of attributes by using Shapley Additive exPlanations (SHAP) values, and exploits kernel extreme learning machine (KELM) due to its nonlinear decision boundary and good generalization ability. However, KELM suffers from the 'curse of dimensionality problem' for large datasets due to the complex computation of kernel matrix with large numbers of hidden nodes. We solved this problem by proposing a novel multicolumn kernel extreme learning machine (MCKELM) which exploited the k-d tree algorithm to divide data into N subsets and trains separate KELM on each subset of data. Then, the method incorporates KELM classifiers into parallel structures and selects the top k nearest subsets during testing by using the k-d tree search for classifying input instead of the whole network. For evaluating a proposed framework large food/non-food dataset is prepared using nine publically available datasets. Experimental results showed the superiority of our method on an integrated set of measures while solving the problem of 'curse of dimensionality in KELM for large datasets.

Learning Car Speed Using Inertial Sensors

A deep neural network (DNN) is trained to estimate the speed of a car driving in an urban area using as input a stream of measurements from a low-cost six-axis inertial measurement unit (IMU). Three hours of data was collected by driving through the city of Ashdod, Israel in a car equipped with a global navigation satellite system (GNSS) real time kinematic (RTK) positioning device and a synchronized IMU. Ground truth labels for the car speed were calculated using the position measurements obtained at the high rate of 50 [Hz]. A DNN architecture with long short-term memory layers is proposed to enable high-frequency speed estimation that accounts for previous inputs history and the nonlinear relation between speed, acceleration, and angular velocity. A simplified aided dead reckoning localization scheme is formulated to assess the trained model which provides the speed pseudo-measurement. The trained model is shown to substantially improve the position accuracy during a 4 minutes drive without the use of GNSS position updates.

Policy Gradient Method For Robust Reinforcement Learning

This paper develops the first policy gradient method with global optimality guarantee and complexity analysis for robust reinforcement learning under model mismatch. Robust reinforcement learning is to learn a policy robust to model mismatch between simulator and real environment. We first develop the robust policy (sub-)gradient, which is applicable for any differentiable parametric policy class. We show that the proposed robust policy gradient method converges to the global optimum asymptotically under direct policy parameterization. We further develop a smoothed robust policy gradient method and show that to achieve an $\epsilon$-global optimum, the complexity is $\mathcal O(\epsilon^{-3})$. We then extend our methodology to the general model-free setting and design the robust actor-critic method with differentiable parametric policy class and value function. We further characterize its asymptotic convergence and sample complexity under the tabular setting. Finally, we provide simulation results to demonstrate the robustness of our methods.

Reductive MDPs: A Perspective Beyond Temporal Horizons

Solving general Markov decision processes (MDPs) is a computationally hard problem. Solving finite-horizon MDPs, on the other hand, is highly tractable with well known polynomial-time algorithms. What drives this extreme disparity, and do problems exist that lie between these diametrically opposed complexities? In this paper we identify and analyse a sub-class of stochastic shortest path problems (SSPs) for general state-action spaces whose dynamics satisfy a particular drift condition. This construction generalises the traditional, temporal notion of a horizon via decreasing reachability: a property called reductivity. It is shown that optimal policies can be recovered in polynomial-time for reductive SSPs -- via an extension of backwards induction -- with an efficient analogue in reductive MDPs. The practical considerations of the proposed approach are discussed, and numerical verification provided on a canonical optimal liquidation problem.

Sobolev Acceleration and Statistical Optimality for Learning Elliptic Equations via Gradient Descent

In this paper, we study the statistical limits in terms of Sobolev norms of gradient descent for solving inverse problem from randomly sampled noisy observations using a general class of objective functions. Our class of objective functions includes Sobolev training for kernel regression, Deep Ritz Methods (DRM), and Physics Informed Neural Networks (PINN) for solving elliptic partial differential equations (PDEs) as special cases. We consider a potentially infinite-dimensional parameterization of our model using a suitable Reproducing Kernel Hilbert Space and a continuous parameterization of problem hardness through the definition of kernel integral operators. We prove that gradient descent over this objective function can also achieve statistical optimality and the optimal number of passes over the data increases with sample size. Based on our theory, we explain an implicit acceleration of using a Sobolev norm as the objective function for training, inferring that the optimal number of epochs of DRM becomes larger than the number of PINN when both the data size and the hardness of tasks increase, although both DRM and PINN can achieve statistical optimality.

Analyzing Lottery Ticket Hypothesis from PAC-Bayesian Theory Perspective

The lottery ticket hypothesis (LTH) has attracted attention because it can explain why over-parameterized models often show high generalization ability. It is known that when we use iterative magnitude pruning (IMP), which is an algorithm to find sparse networks with high generalization ability that can be trained from the initial weights independently, called winning tickets, the initial large learning rate does not work well in deep neural networks such as ResNet. However, since the initial large learning rate generally helps the optimizer to converge to flatter minima, we hypothesize that the winning tickets have relatively sharp minima, which is considered a disadvantage in terms of generalization ability. In this paper, we confirm this hypothesis and show that the PAC-Bayesian theory can provide an explicit understanding of the relationship between LTH and generalization behavior. On the basis of our experimental findings that flatness is useful for improving accuracy and robustness to label noise and that the distance from the initial weights is deeply involved in winning tickets, we offer the PAC-Bayes bound using a spike-and-slab distribution to analyze winning tickets. Finally, we revisit existing algorithms for finding winning tickets from a PAC-Bayesian perspective and provide new insights into these methods.

cMelGAN: An Efficient Conditional Generative Model Based on Mel Spectrograms

Analysing music in the field of machine learning is a very difficult problem with numerous constraints to consider. The nature of audio data, with its very high dimensionality and widely varying scales of structure, is one of the primary reasons why it is so difficult to model. There are many applications of machine learning in music, like the classifying the mood of a piece of music, conditional music generation, or popularity prediction. The goal for this project was to develop a genre-conditional generative model of music based on Mel spectrograms and evaluate its performance by comparing it to existing generative music models that use note-based representations. We initially implemented an autoregressive, RNN-based generative model called MelNet . However, due to its slow speed and low fidelity output, we decided to create a new, fully convolutional architecture that is based on the MelGAN [4] and conditional GAN architectures, called cMelGAN.

Developing patient-driven artificial intelligence based on personal rankings of care decision making steps

We propose and experimentally motivate a new methodology to support decision-making processes in healthcare with artificial intelligence based on personal rankings of care decision making steps that can be identified with our methodology, questionnaire data and its statistical patterns. Our longitudinal quantitative cross-sectional three-stage study gathered self-ratings for 437 expression statements concerning healthcare situations on Likert scales in respect to "the need for help", "the advancement of health", "the hopefulness", "the indication of compassion" and "the health condition", and 45 answers about the person's demographics, health and wellbeing, also the duration of giving answers. Online respondents between 1 June 2020 and 29 June 2021 were recruited from Finnish patient and disabled people's organizations, other health-related organizations and professionals, and educational institutions (n=1075). With Kruskal-Wallis test, Wilcoxon rank-sum test (i.e., Mann-Whitney U test), Wilcoxon rank-sum pairwise test, Welch's t test and one-way analysis of variance (ANOVA) between groups test we identified statistically significant differences of ratings and their durations for each expression statement in respect to respondent groupings based on the answer values of each background question. Frequencies of the later reordering of rating rankings showed dependencies with ratings given earlier in respect to various interpretation task entities, interpretation dimensions and respondent groupings. Our methodology, questionnaire data and its statistical patterns enable analyzing with self-rated expression statements the representations of decision making steps in healthcare situations and their chaining, agglomeration and branching in knowledge entities of personalized care paths. Our results support building artificial intelligence solutions to address the patient's needs concerning care.

Parameter Adaptation for Joint Distribution Shifts

While different methods exist to tackle distinct types of distribution shift, such as label shift (in the form of adversarial attacks) or domain shift, tackling the joint shift setting is still an open problem. Through the study of a joint distribution shift manifesting both adversarial and domain-specific perturbations, we not only show that a joint shift worsens model performance compared to their individual shifts, but that the use of a similar domain worsens performance than a dissimilar domain. To curb the performance drop, we study the use of perturbation sets motivated by input and parameter space bounds, and adopt a meta learning strategy (hypernetworks) to model parameters w.r.t. test-time inputs to recover performance.