Publications

An Analysis of Quantile Temporal-Difference Learning
Mark Rowland
Remi Munos
Mohammad Gheshlaghi Azar
Yunhao Tang
Georg Ostrovski
Anna Harutyunyan
K. Tuyls
Marc Gendron-Bellemare
Will Dabney
We analyse quantile temporal-difference learning (QTD), a distributional reinforcement learning algorithm that has proven to be a key compon… (see more)ent in several successful large-scale applications of reinforcement learning. Despite these empirical successes, a theoretical understanding of QTD has proven elusive until now. Unlike classical TD learning, which can be analysed with standard stochastic approximation tools, QTD updates do not approximate contraction mappings, are highly non-linear, and may have multiple fixed points. The core result of this paper is a proof of convergence to the fixed points of a related family of dynamic programming procedures with probability 1, putting QTD on firm theoretical footing. The proof establishes connections between QTD and non-linear differential inclusions through stochastic approximation theory and non-smooth analysis.
arxiv.org
Assessing Neural Network Representations During Training Using Noise-Resilient Diffusion Spectral Entropy
Danqi Liao
Chen Liu
Benjamin W Christensen
Alexander Tong
Guillaume Huguet
Guy Wolf
Maximilian Nickel
Ian Adelstein
Smita Krishnaswamy
Entropy and mutual information in neural networks provide rich information on the learning process, but they have proven difficult to comput… (see more)e reliably in high dimensions. Indeed, in noisy and high-dimensional data, traditional estimates in ambient dimensions approach a fixed entropy and are prohibitively hard to compute. To address these issues, we leverage data geometry to access the underlying manifold and reliably compute these information-theoretic measures. Specifically, we define diffusion spectral entropy (DSE) in neural representations of a dataset as well as diffusion spectral mutual information (DSMI) between different variables representing data. First, we show that they form noise-resistant measures of intrinsic dimensionality and relationship strength in high-dimensional simulated data that outperform classic Shannon entropy, nonparametric estimation, and mutual information neural estimation (MINE). We then study the evolution of representations in classification networks with supervised learning, self-supervision, or overfitting. We observe that (1) DSE of neural representations increases during training; (2) DSMI with the class label increases during generalizable learning but stays stagnant during overfitting; (3) DSMI with the input signal shows differing trends: on MNIST it increases, while on CIFAR-10 and STL-10 it decreases. Finally, we show that DSE can be used to guide better network initialization and that DSMI can be used to predict downstream classification accuracy across 962 models on ImageNet.
2024-01-01
CISS (published)
doi.org
openreview.net
BAND: Biomedical Alert News Dataset
Zihao Fu
Meiru Zhang
Zaiqiao Meng
Yannan Shen
Anya Okhmatovskaia
David Buckeridge
Nigel Collier
2024-01-01
AAAI (published)
doi.org
arxiv.org
A benchmark of individual auto-regressive models in a massive fMRI dataset
Fraçois Paugam
Basile Pinsard
Guillaume Lajoie
Pierre (Louis) Bellec
Dense functional magnetic resonance imaging datasets open new avenues to create auto-regressive models of brain activity. Individual idiosyn… (see more)crasies are obscured by group models, but can be captured by purely individual models given sufficient amounts of training data. In this study, we compared several deep and shallow individual models on the temporal auto-regression of BOLD time series recorded during a natural video watching task. The best performing models were then analyzed in terms of their data requirements and scaling, subject specificity and the space-time structure of their predicted dynamics. We found the Chebnets, a type of graph convolutional neural network, to be best suited for temporal BOLD auto-regression, closely followed by linear models. Chebnets demonstrated an increase in performance with increasing amounts of data, with no complete saturation at 9 h of training data. Good generalization to other kinds of video stimuli and to resting state data marked the Chebnets’ ability to capture intrinsic brain dynamics rather than only stimulus-specific autocorrelation patterns. Significant subject specificity was found at short prediction time lags. The Chebnets were found to capture lower frequencies at longer prediction time lags, and the spatial correlations in predicted dynamics were found to match traditional functional connectivity networks. Overall, these results demonstrate that large individual fMRI datasets can be used to efficiently train purely individual auto-regressive models of brain activity, and that massive amounts of individual data are required to do so. The excellent performance of the Chebnets likely reflects their ability to combine spatial and temporal interactions on large time scales at a low complexity cost. The non-linearities of the models did not appear as a key advantage. In fact, surprisingly, linear versions of the Chebnets appeared to outperform the original nonlinear ones. Individual temporal auto-regressive models have the potential to improve the predictability of the BOLD signal. This study is based on a massive, publicly-available dataset, which can serve for future benchmarks of individual auto-regressive modeling.
2024-01-01
Imaging Neuroscience (published)
doi.org
BETAC: Bidirectional Encoder Transformer for Assembly Code Function Name Recovery
Guillaume Breyton
Mohd Saqib
Benjamin Fung
Philippe Charland
Recovering function names from stripped binaries is a crucial and time-consuming task for software reverse engineering’ particularly in en… (see more)hancing network reliability, resilience, and security. This paper tackles the challenge of recovering function names in stripped binaries, a fundamental step in reverse engineering. The absence of syntactic information and the possibility of different code producing identical behavior complicate this task. To overcome these challenges, we introduce a novel model, the Bidirectional Encoder Transformer for Assembly Code (BETAC), leveraging a transformer-based architecture known for effectively processing sequential data. BETAC utilizes self-attention mechanisms and feed-forward networks to discern complex relationships within assembly code for precise function name prediction. We evaluated BETAC against various existing encoder and decoder models in diverse binary datasets, including benign and malicious codes in multiple formats. Our model demonstrated superior performance over previous techniques in certain metrics and showed resilience against code obfuscation.
2024-01-01
DRCN (published)
doi.org
Carbon capture, utilization and sequestration systems design and operation optimization: Assessment and perspectives of artificial intelligence opportunities
Eslam G. Al-Sakkari
Ahmed Ragab
Hanane Dagdougui
Daria Camilla Boffito
Mouloud Amazouz
2024-01-01
Science of the Total Environment (published)
doi.org
Causal Adversarial Perturbations for Individual Fairness and Robustness in Heterogeneous Data Spaces
Ahmad-reza Ehyaei
Kiarash Mohammadi
Amir-Hossein Karimi
S. Samadi
Golnoosh Farnadi
2024-01-01
AAAI (published)
doi.org
arxiv.org
Common Challenges of Deep Reinforcement Learning Applications Development: An Empirical Study
Mohammad Mehdi Morovati
Florian Tambon
Mina Taraghi
Amin Nikanjam
Foutse Khomh
2024-01-01
Empir. Softw. Eng. (published)
doi.org
arxiv.org
Connecting Weighted Automata, Tensor Networks and Recurrent Neural Networks through Spectral Learning
Tianyu Li
Doina Precup
Guillaume Rabusseau
2024-01-01
Mach. Learn. (published)
doi.org
arxiv.org
Consolidating Separate Degradations Model via Weights Fusion and Distillation
Dinesh Daultani
Hugo Larochelle
Real-world images prevalently contain different varieties of degradation, such as motion blur and luminance noise. Computer vision recogniti… (see more)on models trained on clean images perform poorly on degraded images. Previously, several works have explored how to perform image classification of degraded images while training a single model for each degradation. Nevertheless, it becomes challenging to host several degradation models for each degradation on limited hardware applications and to estimate degradation parameters correctly at the run-time. This work proposes a method for effectively combining several models trained separately on different degradations into a single model to classify images with different types of degradations. Our proposed method is four-fold: (1) train a base model on clean images, (2) fine-tune the base model in-dividually for all given image degradations, (3) perform a fusion of weights given the fine-tuned models for individual degradations, (4) perform fine-tuning on given task using distillation and cross-entropy loss. Our proposed method can outperform previous state-of-the-art methods of pretraining in out-of-distribution generalization based on degradations such as JPEG compression, salt-and-pepper noise, Gaussian blur, and additive white Gaussian noise by 2.5% on CIFAR-100 dataset and by 1.3% on CIFAR-10 dataset. Moreover, our proposed method can handle degra-dation used for training without any explicit information about degradation at the inference time. Code will be available at https://github.com/dineshdaultani/FusionDistill.
2024-01-01
2024 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) (published)
doi.org
Decoding of Polar Codes Using Quadratic Unconstrained Binary Optimization
Huayi Zhou
Ryan Seah
Marwan Jalaleddine
Warren Gross
2024-01-01
IEEE Journal on Selected Areas in Communications (published)
doi.org
Deep reinforcement learning for continuous wood drying production line control
François-Alexandre Tremblay
Audrey Durand
Michael Morin
Philippe Marier
Jonathan Gaudreault
2024-01-01
Computers in Industry (published)
doi.org