Portrait de Pablo Samuel Castro

Pablo Samuel Castro

Membre industriel principal
Professeur associé, Université de Montréal, Département d'informatique et de recherche opérationnelle
Chercheur scientifique, Google DeepMind
Sujets de recherche
Apprentissage par renforcement
Site web
Google Scholar
LinkedIn
X

Biographie

Pablo Samuel Castro est né et a grandi à Quito, en Équateur, et a déménagé à Montréal après l'école secondaire pour étudier à l’Université McGill. Il y a obtenu un doctorat en se concentrant sur l'apprentissage par renforcement, sous la supervision de Doina Precup et Prakash Panangaden. Il est chercheur scientifique à Google DeepMind à Montréal. Il s’intéresse particulièrement à la recherche fondamentale sur l'apprentissage par renforcement et plaide régulièrement en faveur d'une augmentation de la représentation des personnes d’origine latino-américaine dans la communauté de recherche. Il est également professeur adjoint au Département d'informatique et de recherche opérationnelle (DIRO) de l'Université de Montréal. Outre son intérêt pour le codage, l'intelligence artificielle et les mathématiques, Pablo Samuel est un musicien actif.

Étudiants actuels

Roger Creus-Castanyer
Doctorat - UdeM
Superviseur⋅e principal⋅e :
Glen Berseth
Site web
Github
Google Scholar
Julian Dierkes
Visiteur de recherche indépendant - RWTH Aachen University
Site web
Github
Google Scholar
Nicola Gauthier
Maîtrise recherche - UdeM
Olya Mastikhina
Doctorat - UdeM
Walter Mayor
Collaborateur·rice de recherche
Site web
Github
Google Scholar
Johan Samir Obando Ceron
Doctorat - UdeM
Superviseur⋅e principal⋅e :
Aaron Courville
Site web
Github
Google Scholar
Gandharv Patil
Doctorat - McGill
Superviseur⋅e principal⋅e :
Doina Precup
Site web
Github
Google Scholar
Ali Saheb Pasand
Doctorat - McGill
Superviseur⋅e principal⋅e :
Blake Richards
Dhruv Sreenivas
Doctorat - UdeM
Site web
Github

Publications

Discovering Symbolic Cognitive Models from Human and Animal Behavior
Pablo Samuel Castro
Nenad Tomasev
Ankit Anand
Navodita Sharma
Rishika Mohanta
Aparna Dev
Kuba Perlin
Siddhant Jain
Kyle Levin
Noemi Elteto
Will Dabney
Alexander Novikov
Glenn C Turner
Maria K Eckstein
Nathaniel D. Daw
Kevin J Miller
Kim Stachenfeld
Symbolic models play a key role in cognitive science, expressing computationally precise hypotheses about how the brain implements a cogniti… (voir plus)ve process. Identifying an appropriate model typically requires a great deal of effort and ingenuity on the part of a human scientist. Here, we adapt FunSearch (Romera-Paredes et al. 2024), a recently developed tool that uses Large Language Models (LLMs) in an evolutionary algorithm, to automatically discover symbolic cognitive models that accurately capture human and animal behavior. We consider datasets from three species performing a classic reward-learning task that has been the focus of substantial modeling effort, and find that the discovered programs outperform state-of-the-art cognitive models for each. The discovered programs can readily be interpreted as hypotheses about human and animal cognition, instantiating interpretable symbolic learning and decision-making algorithms. Broadly, these results demonstrate the viability of using LLM-powered program synthesis to propose novel scientific hypotheses regarding mechanisms of human and animal cognition.
2025-04-30
ICML.cc/2025/Conference (poster)
proceedings.mlr.press
proceedings.mlr.press
Mitigating Plasticity Loss in Continual Reinforcement Learning by Reducing Churn
Hongyao Tang
Johan Obando-Ceron
Pablo Samuel Castro
Aaron Courville
Glen Berseth
Plasticity, or the ability of an agent to adapt to new tasks, environments, or distributions, is crucial for continual learning. In this pap… (voir plus)er, we study the loss of plasticity in deep continual RL from the lens of churn: network output variability for out-of-batch data induced by mini-batch training. We demonstrate that (1) the loss of plasticity is accompanied by the exacerbation of churn due to the gradual rank decrease of the Neural Tangent Kernel (NTK) matrix; (2) reducing churn helps prevent rank collapse and adjusts the step size of regular RL gradients adaptively. Moreover, we introduce Continual Churn Approximated Reduction (C-CHAIN) and demonstrate it improves learning performance and outperforms baselines in a diverse range of continual learning environments on OpenAI Gym Control, ProcGen, DeepMind Control Suite, and MinAtar benchmarks.
2025-04-30
International Conference on Machine Learning (poster)
doi.org
proceedings.mlr.press
Don't Flatten, Tokenize! Unlocking the Key to SoftMoE's Efficacy in Deep RL
Ghada Sokar
Johan Obando-Ceron
Aaron Courville
Hugo Larochelle
Pablo Samuel Castro
The use of deep neural networks in reinforcement learning (RL) often suffers from performance degradation as model size increases. While sof… (voir plus)t mixtures of experts (SoftMoEs) have recently shown promise in mitigating this issue for online RL, the reasons behind their effectiveness remain largely unknown. In this work we provide an in-depth analysis identifying the key factors driving this performance gain. We discover the surprising result that tokenizing the encoder output, rather than the use of multiple experts, is what is behind the efficacy of SoftMoEs. Indeed, we demonstrate that even with an appropriately scaled single expert, we are able to maintain the performance gains, largely thanks to tokenization.
2025-01-21
ICLR.cc/2025/Conference (spotlight)
doi.org
openreview.net
Studying the Interplay Between the Actor and Critic Representations in Reinforcement Learning
Samuel Garcin
Trevor McInroe
Pablo Samuel Castro
Prakash Panangaden
Christopher G. Lucas
David Abel
Stefano V Albrecht
2025-01-21
ICLR.cc/2025/Conference (poster)
doi.org
openreview.net
Overcoming State and Action Space Disparities in Multi-Domain, Multi-Task Reinforcement Learning
Reginald McLean
Kai Yuan
Isaac Woungang
Nariman Farsad
Pablo Samuel Castro
Current multi-task reinforcement learning (MTRL) methods have the ability to perform a large number of tasks with a single policy. However w… (voir plus)hen attempting to interact with a new domain, the MTRL agent would need to be re-trained due to differences in domain dynamics and structure. Because of these limitations, we are forced to train multiple policies even though tasks may have shared dynamics, leading to needing more samples and is thus sample inefficient. In this work, we explore the ability of MTRL agents to learn in various domains with various dynamics by simultaneously learning in multiple domains, without the need to fine-tune extra policies. In doing so we find that a MTRL agent trained in multiple domains induces an increase in sample efficiency of up to 70\% while maintaining the overall success rate of the MTRL agent.
2024-10-22
corl.org/2024/Workshop/MAPoDeL (publié)
openreview.net
openreview.net
CALE: Continuous Arcade Learning Environment
Jesse Farebrother
Pablo Samuel Castro
We introduce the Continuous Arcade Learning Environment (CALE), an extension of the well-known Arcade Learning Environment (ALE) [Bellemare … (voir plus)et al., 2013]. The CALE uses the same underlying emulator of the Atari 2600 gaming system (Stella), but adds support for continuous actions. This enables the benchmarking and evaluation of continuous-control agents (such as PPO [Schulman et al., 2017] and SAC [Haarnoja et al., 2018]) and value-based agents (such as DQN [Mnih et al., 2015] and Rainbow [Hessel et al., 2018]) on the same environment suite. We provide a series of open questions and research directions that CALE enables, as well as initial baseline results using Soft Actor-Critic. CALE is available as part of the ALE athttps://github.com/Farama-Foundation/Arcade-Learning-Environment.
2024-09-25
NeurIPS.cc/2024/Datasets_and_Benchmarks_Track (poster)
doi.org
openreview.net
In value-based deep reinforcement learning, a pruned network is a good network
Johan Obando-Ceron
Aaron Courville
Pablo Samuel Castro
Recent work has shown that deep reinforcement learning agents have difficulty in effectively using their network parameters. We leverage pri… (voir plus)or insights into the advantages of sparse training techniques and demonstrate that gradual magnitude pruning enables value-based agents to maximize parameter effectiveness. This results in networks that yield dramatic performance improvements over traditional networks, using only a small fraction of the full network parameters.
2024-07-22
ICML (Accept (Poster))
doi.org
proceedings.mlr.press
Adaptive Accompaniment with ReaLchords
Yusong Wu
Tim Cooijmans
Kyle Kastner
Adam Roberts
Ian Simon
Alexander Scarlatos
Chris Donahue
Cassie Tarakajian
Shayegan Omidshafiei
Aaron Courville
Pablo Samuel Castro
Natasha Jaques
Cheng-Zhi Anna Huang
Jamming requires coordination, anticipation, and collaborative creativity between musicians. Current generative models of music produce expr… (voir plus)essive output but are not able to generate in an \emph{online} manner, meaning simultaneously with other musicians (human or otherwise). We propose ReaLchords, an online generative model for improvising chord accompaniment to user melody. We start with an online model pretrained by maximum likelihood, and use reinforcement learning to finetune the model for online use. The finetuning objective leverages both a novel reward model that provides feedback on both harmonic and temporal coherency between melody and chord, and a divergence term that implements a novel type of distillation from a teacher model that can see the future melody. Through quantitative experiments and listening tests, we demonstrate that the resulting model adapts well to unfamiliar input and produce fitting accompaniment. ReaLchords opens the door to live jamming, as well as simultaneous co-creation in other modalities.
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
proceedings.mlr.press
Stop Regressing: Training Value Functions via Classification for Scalable Deep RL
Jesse Farebrother
Jordi Orbay
Quan Vuong
Adrien Ali Taiga
Yevgen Chebotar
Ted Xiao
Alex Irpan
Sergey Levine
Pablo Samuel Castro
Aleksandra Faust
Aviral Kumar
Rishabh Agarwal
Value functions are a central component of deep reinforcement learning (RL). These functions, parameterized by neural networks, are trained … (voir plus)using a mean squared error regression objective to match bootstrapped target values. However, scaling value-based RL methods that use regression to large networks, such as high-capacity Transformers, has proven challenging. This difficulty is in stark contrast to supervised learning: by leveraging a cross-entropy classification loss, supervised methods have scaled reliably to massive networks. Observing this discrepancy, in this paper, we investigate whether the scalability of deep RL can also be improved simply by using classification in place of regression for training value functions. We demonstrate that value functions trained with categorical cross-entropy significantly improves performance and scalability in a variety of domains. These include: single-task RL on Atari 2600 games with SoftMoEs, multi-task RL on Atari with large-scale ResNets, robotic manipulation with Q-transformers, playing Chess without search, and a language-agent Wordle task with high-capacity Transformers, achieving state-of-the-art results on these domains. Through careful analysis, we show that the benefits of categorical cross-entropy primarily stem from its ability to mitigate issues inherent to value-based RL, such as noisy targets and non-stationarity. Overall, we argue that a simple shift to training value functions with categorical cross-entropy can yield substantial improvements in the scalability of deep RL at little-to-no cost.
2024-07-07
Proceedings of the 41st International Conference on Machine Learning (publié)
doi.org
proceedings.mlr.press
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 D. Cubuk
Aaron Courville
Bellemare Marc-Emmanuel
Joshua Robinson
Pablo Samuel Castro
Sergei V Kalinin
2024-06-30
Microscopy and Microanalysis (inconnu)
doi.org
Mixtures of Experts Unlock Parameter Scaling for Deep RL
Johan Obando-Ceron
Ghada Sokar
Timon Willi
Clare Lyle
Jesse Farebrother
Jakob Foerster
Karolina Dziugaite
Doina Precup
Pablo Samuel Castro
The recent rapid progress in (self) supervised learning models is in large part predicted by empirical scaling laws: a model's performance s… (voir plus)cales proportionally to its size. Analogous scaling laws remain elusive for reinforcement learning domains, however, where increasing the parameter count of a model often hurts its final performance. In this paper, we demonstrate that incorporating Mixture-of-Expert (MoE) modules, and in particular Soft MoEs (Puigcerver et al., 2023), into value-based networks results in more parameter-scalable models, evidenced by substantial performance increases across a variety of training regimes and model sizes. This work thus provides strong empirical evidence towards developing scaling laws for reinforcement learning.
2024-04-30
ICML.cc/2024/Conference (spotlight)
doi.org
proceedings.mlr.press
A density estimation perspective on learning from pairwise human preferences
Vincent Dumoulin
Daniel D. Johnson
Pablo Samuel Castro
Hugo Larochelle
Yann Dauphin
Learning from human feedback (LHF) -- and in particular learning from pairwise preferences -- has recently become a crucial ingredient in tr… (voir plus)aining large language models (LLMs), and has been the subject of much research. Most recent works frame it as a reinforcement learning problem, where a reward function is learned from pairwise preference data and the LLM is treated as a policy which is adapted to maximize the rewards, often under additional regularization constraints. We propose an alternative interpretation which centers on the generative process for pairwise preferences and treats LHF as a density estimation problem. We provide theoretical and empirical results showing that for a family of generative processes defined via preference behavior distribution equations, training a reward function on pairwise preferences effectively models an annotator's implicit preference distribution. Finally, we discuss and present findings on"annotator misspecification"-- failure cases where wrong modeling assumptions are made about annotator behavior, resulting in poorly-adapted models -- suggesting that approaches that learn from pairwise human preferences could have trouble learning from a population of annotators with diverse viewpoints.
2024-02-26
TMLR (accepté)
doi.org
openreview.net