Publications

Normalization and effective learning rates in reinforcement learning
Clare Lyle
Zeyu Zheng
Khimya Khetarpal
James Martens
Hado van Hasselt
Razvan Pascanu
Will Dabney
Normalization layers have recently experienced a renaissance in the deep reinforcement learning and continual learning literature, with seve… (see more)ral works highlighting diverse benefits such as improving loss landscape conditioning and combatting overestimation bias. However, normalization brings with it a subtle but important side effect: an equivalence between growth in the norm of the network parameters and decay in the effective learning rate. This becomes problematic in continual learning settings, where the resulting effective learning rate schedule may decay to near zero too quickly relative to the timescale of the learning problem. We propose to make the learning rate schedule explicit with a simple re-parameterization which we call Normalize-and-Project (NaP), which couples the insertion of normalization layers with weight projection, ensuring that the effective learning rate remains constant throughout training. This technique reveals itself as a powerful analytical tool to better understand learning rate schedules in deep reinforcement learning, and as a means of improving robustness to nonstationarity in synthetic plasticity loss benchmarks along with both the single-task and sequential variants of the Arcade Learning Environment. We also show that our approach can be easily applied to popular architectures such as ResNets and transformers while recovering and in some cases even slightly improving the performance of the base model in common stationary benchmarks.
2024-07-01
ArXiv (preprint)
doi.org
arxiv.org
One-shot Learning for MIPs with SOS1 Constraints
Charly Robinson La Rocca
Jean-François Cordeau
Emma Frejinger
2024-07-01
Operations Research Forum (published)
doi.org
arxiv.org
Surface water temperature observations and ice phenology estimations for 1.4 million lakes globally
Maartje C. Korver
Bernhard Lehner
Jeff Cardille
Laura Carrea
2024-07-01
Remote Sensing of Environment (published)
doi.org
The Position Dependence of Electron Beam Induced Effects in 2D Materials with Deep Neural Networks
Kevin M Roccapriore
Max Schwarzer
Joshua Greaves
Jesse Farebrother
Riccardo Torsi
Rishabh Agarwal
Colton Bishop
Igor Mordatch
Ekin Dogus Cubuk
Aaron Courville
Marc Gendron-Bellemare
Joshua Robinson
Pablo Samuel Castro
Sergei V Kalinin
2024-07-01
Microscopy and Microanalysis (published)
doi.org
Mirror Descent Algorithms with Nearly Dimension-Independent Rates for Differentially-Private Stochastic Saddle-Point Problems extended abstract
Tomas Gonzalez
Cristobal Guzman
Courtney Paquette
2024-06-30
Proceedings of Thirty Seventh Conference on Learning Theory (published)
proceedings.mlr.press
Open-Source Conversational AI with SpeechBrain 1.0
Mirco Ravanelli
Titouan Parcollet
Adel Moumen
Sylvain de Langen
Cem Subakan
Peter William VanHarn Plantinga
Yingzhi Wang
Pooneh Mousavi
Luca Della Libera
Artem Ploujnikov
Francesco Paissan
Davide Borra
Salah Zaiem
Zeyu Zhao
Shucong Zhang
Georgios Karakasidis
Sung-Lin Yeh
Pierre Champion
Aku Rouhe
Rudolf Braun … (see 11 more)
Florian Mai
Juan Pablo Zuluaga
Seyed Mahed Mousavi
Andreas Nautsch
Xuechen Liu
Sangeet Sagar
Jarod Duret
Salima Mdhaffar
G. Laperriere
Renato de Mori
Yannick Estève
SpeechBrain is an open-source Conversational AI toolkit based on PyTorch, focused particularly on speech processing tasks such as speech rec… (see more)ognition, speech enhancement, speaker recognition, text-to-speech, and much more. It promotes transparency and replicability by releasing both the pre-trained models and the complete"recipes"of code and algorithms required for training them. This paper presents SpeechBrain 1.0, a significant milestone in the evolution of the toolkit, which now has over 200 recipes for speech, audio, and language processing tasks, and more than 100 models available on Hugging Face. SpeechBrain 1.0 introduces new technologies to support diverse learning modalities, Large Language Model (LLM) integration, and advanced decoding strategies, along with novel models, tasks, and modalities. It also includes a new benchmark repository, offering researchers a unified platform for evaluating models across diverse tasks
2024-06-29
ArXiv (preprint)
doi.org
arxiv.org
Variable Time Step Reinforcement Learning for Robotic Applications
Dong Wang
Giovanni Beltrame
Traditional reinforcement learning (RL) generates discrete control policies, assigning one action per cycle. These policies are usually impl… (see more)emented as in a fixed-frequency control loop. This rigidity presents challenges as optimal control frequency is task-dependent; suboptimal frequencies increase computational demands and reduce exploration efficiency. Variable Time Step Reinforcement Learning (VTS-RL) addresses these issues with adaptive control frequencies, executing actions only when necessary, thus reducing computational load and extending the action space to include action durations. In this paper we introduce the Multi-Objective Soft Elastic Actor-Critic (MOSEAC) method to perform VTS-RL, validating it through theoretical analysis and experimentation in simulation and on real robots. Results show faster convergence, better training results, and reduced energy consumption with respect to other variable- or fixed-frequency approaches.
2024-06-29
ArXiv (preprint)
doi.org
arxiv.org
Adversarial Training with Synthesized Data: A Path to Robust and Generalizable Neural Networks
Reza Bayat
Irina Rish
Adversarial Training (AT) is a well-known framework designed to mitigate adversarial vulnerabilities in neural networks. Recent research ind… (see more)icates that incorporating adversarial examples (AEs) in training can enhance models' generalization capabilities. To understand the impact of AEs on learning dynamics, we study AT through the lens of sample difficulty methodologies. Our findings show that AT leads to more stable learning dynamics compared to Natural Training (NT), resulting in gradual performance improvements and less overconfident predictions. This suggests that AT steers training away from learning easy, perturbable spurious features toward more resilient and generalizable ones. However, a trade-off exists between adversarial robustness and generalization gains, due to robust overfitting, limiting practical deployment. To address this, we propose using synthesized data to bridge this gap. Our results demonstrate that AT benefits significantly from synthesized data, whereas NT does not, enhancing generalization without compromising robustness and offering new avenues for developing robust and generalizable models.
2024-06-28
ICML.cc/2024/Workshop/NextGenAISafety (poster)
openreview.net
openreview.net
Decomposed evaluations of geographic disparities in text-to-image models
Abhishek Sureddy
Dishant Padalia
Nandhinee Periyakaruppan
Oindrila Saha
Adina Williams
Adriana Romero Soriano
Megan Richards
Polina Kirichenko
Melissa Hall
2024-06-28
ICML.cc/2024/Workshop/NextGenAISafety (poster)
doi.org
openreview.net
Economic evaluation of the effect of needle and syringe programs on skin, soft tissue, and vascular infections in people who inject drugs: a microsimulation modelling approach
Jihoon Lim
W Alton Russell
Mariam El-Sheikh
David Buckeridge
Dimitra Panagiotoglou
2024-06-28
Harm Reduction Journal (published)
doi.org
Exploring Scaling Trends in LLM Robustness
Nikolaus H. R. Howe
Michał Zając
Ian R. McKenzie
Oskar John Hollinsworth
Tom Tseng
Aaron David Tucker
Pierre-Luc Bacon
Adam Gleave
Language model capabilities predictably improve from scaling a model's size and training data. Motivated by this, increasingly large languag… (see more)e models have been trained, yielding an array of impressive capabilities. Yet these models are vulnerable to adversarial prompts, such as"jailbreaks"that hijack models to perform undesired behaviors, posing a significant risk of misuse. Prior work indicates that computer vision models become more robust with model and data scaling, raising the question: does language model robustness also improve with scale? We study this question empirically, finding that larger models respond substantially better to adversarial training, but there is little to no benefit from model scale in the absence of explicit defenses.
2024-06-28
ICML.cc/2024/Workshop/NextGenAISafety (poster)
doi.org
openreview.net
Game On, Hate Off: A Study of Toxicity in Online Multiplayer Environments
Zachary Yang
Nicolas Grenon-Godbout
Reihaneh Rabbany
2024-06-28
Games: Research and Practice (published)
doi.org