Doina Precup

Sumana Basu

Collaborateur·rice alumni - McGill

Co-superviseur⋅e :

Adriana Romero Soriano

Collaborateur·rice alumni - McGill

Lynn Cherif

Collaborateur·rice alumni - McGill

Co-superviseur⋅e :

Doctorat - McGill

Co-superviseur⋅e :

Doctorat - McGill

Superviseur⋅e principal⋅e :

David Meger

Jonathan Colaço Carr

Maîtrise recherche - McGill

Superviseur⋅e principal⋅e :

Prakash Panangaden

Élodie Coté-Gauthier

Collaborateur·rice de recherche - McGill

Co-superviseur⋅e :

Isabeau Prémont-Schwarz

Franco Del Balso

Collaborateur·rice de recherche - UdeM

Jesse Farebrother

Doctorat - McGill

Superviseur⋅e principal⋅e :

Marc Gendron-Bellemare

Doctorat - McGill

Superviseur⋅e principal⋅e :

Collaborateur·rice de recherche - Birla Institute of Technology

Howard Huang

Doctorat - McGill

Haque Ishfaq

Collaborateur·rice alumni - McGill

Mohammad Sami Nur Islam Islam

Maîtrise recherche - McGill

Arushi Jain

Collaborateur·rice alumni - McGill

Doctorat - Polytechnique

Doctorat - McGill

Postdoctorat - McGill

Jonathan Lebensold

Collaborateur·rice alumni - McGill

Collaborateur·rice alumni - McGill

Ray Luo

Doctorat - McGill

Superviseur⋅e principal⋅e :

G McCracken

Doctorat - McGill

Nazanin Mohammadi Sepahvand

Collaborateur·rice alumni - McGill

Shahrad Mohammadzadeh

Maîtrise recherche - McGill

Superviseur⋅e principal⋅e :

Gabriela Moisescu-Pareja

Collaborateur·rice de recherche - McGill

Co-superviseur⋅e :

Irina Rish

Padideh Nouri

Doctorat - UdeM

Co-superviseur⋅e :

Doctorat - McGill

Co-superviseur⋅e :

Nate Rahn

Doctorat - McGill

Superviseur⋅e principal⋅e :

Marc Gendron-Bellemare

Sahand Rezaei-Shoshtari

Doctorat - McGill

Co-superviseur⋅e :

Doctorat - McGill

Co-superviseur⋅e :

Doctorat - McGill

Co-superviseur⋅e :

Doctorat - McGill

Nishanth Anand Vemgal

Doctorat - McGill

Doctorat - McGill

Co-superviseur⋅e :

Samira Ebrahimi Kahou

Zihan Wang

Doctorat - McGill

Skipper : combiner l’abstraction spatiale et temporelle afin d’améliorer la généralisation

Guangyuan Wang

Stagiaire de recherche - McGill

Steve Wen

Maîtrise recherche - McGill

Co-superviseur⋅e :

Gregory Dudek

Zijing Wu

Doctorat - McGill

Superviseur⋅e principal⋅e :

Doctorat - McGill

Harry Zhao

Collaborateur·rice alumni - McGill

Co-superviseur⋅e :

Billets de blogue

Generic thumbnail for Mila Blog articles.

22 février 2024

par

Mingde Harry Zhao

Safa Alver

Harm van Seijen

Romain Laroche

Doina Precup

Yoshua Bengio

Lire l'article

Publications

Adaptive Exploration for Data-Efficient General Value Function Evaluations

Arushi Jain

Josiah P. Hanna

General Value Functions (GVFs) (Sutton et al., 2011) represent predictive knowledge in reinforcement learning. Each GVF computes the expecte… (voir plus)d return for a given policy, based on a unique reward. Existing methods relying on fixed behavior policies or pre-collected data often face data efficiency issues when learning multiple GVFs in parallel using off-policy methods. To address this, we introduce GVFExplorer, which adaptively learns a single behavior policy that efficiently collects data for evaluating multiple GVFs in parallel. Our method optimizes the behavior policy by minimizing the total variance in return across GVFs, thereby reducing the required environmental interactions. We use an existing temporal-difference-style variance estimator to approximate the return variance. We prove that each behavior policy update decreases the overall mean squared error in GVF predictions. We empirically show our method's performance in tabular and nonlinear function approximation settings, including Mujoco environments, with stationary and non-stationary reward signals, optimizing data usage and reducing prediction errors across multiple GVFs.

2024-09-24

NeurIPS.cc/2024/Conference (poster)

Parseval Regularization for Continual Reinforcement Learning

Loss of plasticity, trainability loss, and primacy bias have been identified as issues arising when training deep neural networks on sequenc… (voir plus)es of tasks -- all referring to the increased difficulty in training on new tasks. We propose to use Parseval regularization, which maintains orthogonality of weight matrices, to preserve useful optimization properties and improve training in a continual reinforcement learning setting. We show that it provides significant benefits to RL agents on a suite of gridworld, CARL and MetaWorld tasks. We conduct comprehensive ablations to identify the source of its benefits and investigate the effect of certain metrics associated to network trainability including weight matrix rank, weight norms and policy entropy.

2024-09-24

NeurIPS.cc/2024/Conference (poster)

QGFN: Controllable Greediness with Action Values

Stephen Zhewen Lu

Generative Flow Networks (GFlowNets; GFNs) are a family of energy-based generative methods for combinatorial objects, capable of generating … (voir plus)diverse and high-utility samples. However, consistently biasing GFNs towards producing high-utility samples is non-trivial. In this work, we leverage connections between GFNs and reinforcement learning (RL) and propose to combine the GFN policy with an action-value estimate,

2024-09-24

NeurIPS.cc/2024/Conference (poster)

Understanding Decision-Time vs. Background Planning in Model-Based Reinforcement Learning

Safa Alver

In model-based reinforcement learning, an agent can leverage a learned model to improve its way of behaving in different ways. Two prevalent… (voir plus) approaches are decision-time planning and background planning. In this study, we are interested in understanding under what conditions and in which settings one of these two planning styles will perform better than the other in domains that require fast responses. After viewing them through the lens of dynamic programming, we first consider the classical instantiations of these planning styles and provide theoretical results and hypotheses on which one will perform better in the pure planning, planning&learning, and transfer learning settings. We then consider the modern instantiations of these planning styles and provide hypotheses on which one will perform better in the last two of the considered settings. Lastly, we perform several illustrative experiments to empirically validate both our theoretical results and hypotheses. Overall, our findings suggest that even though decision-time planning does not perform as well as background planning in their classical instantiations, in their modern instantiations, it can perform on par or better than background planning in both the planning&learning and transfer learning settings.

2024-07-31

EWRL/2024/Workshop (accepté)

Functional Acceleration for Policy Mirror Descent

Veronica Chelu

2024-06-18

ICML.cc/2024/Workshop/ARLET (poster)

Recurrent Policies Are Not Enough for Continual Reinforcement Learning

Nathan Samuel de Lara

Veronica Chelu

Continual Reinforcement Learning (CRL) aims to develop algorithms that adapt to non-stationary sequences of tasks. A promising recent approa… (voir plus)ch utilizes Recurrent Neural Networks (RNNs) to learn contextual Markov Decision Process (MDP) embeddings. This enables a reinforcement learning (RL) agent to discern the optimality of actions across diverse tasks. In this study, we examine two critical failure modes in the learning of these contextual MDP embeddings. Specifically, we find that RNNs are prone to catastrophic forgetting, manifesting in two distinct ways: (i) embedding collapse---where agents initially learn a contextual task structure that later collapses to a single task, and (ii) embedding drift---where learning embeddings for new MDPs interferes with embeddings the RNN outputs for previous MDPs in the sequence, leading to suboptimal performance of downstream policy networks conditioned on stale embeddings. We explore the effects of various objective functions and network architectures concerning these failure modes, revealing that one of these modes consistently emerges across different setups.

2024-06-06

rl-conference.cc/RLC/2024/Workshop/ICBINB (poster)

On the Limits of Multi-modal Meta-Learning with Auxiliary Task Modulation Using Conditional Batch Normalization

Jordi Armengol-Estap'e

Vincent Michalski

Ramnath Kumar

Pierre-Luc St-Charles

S Ebrahimi Kahou

Few-shot learning aims to learn representations that can tackle novel tasks given a small number of examples. Recent studies show that cross… (voir plus)-modal learning can improve representations for few-shot classification. More specifically, language is a rich modality that can be used to guide visual learning. In this work, we experiment with a multi-modal architecture for few-shot learning that consists of three components: a classifier, an auxiliary network, and a bridge network. While the classifier performs the main classification task, the auxiliary network learns to predict language representations from the same input, and the bridge network transforms high-level features of the auxiliary network into modulation parameters for layers of the few-shot classifier using conditional batch normalization. The bridge should encourage a form of lightweight semantic alignment between language and vision which could be useful for the classifier. However, after evaluating the proposed approach on two popular few-shot classification benchmarks we find that a) the improvements do not reproduce across benchmarks, and b) when they do, the improvements are due to the additional compute and parameters introduced by the bridge network. We contribute insights and recommendations for future work in multi-modal meta-learning, especially when using language representations.

2024-05-28

ArXiv (prépublication)

arxiv.org

More Efficient Randomized Exploration for Reinforcement Learning via Approximate Sampling

Haque Ishfaq

Yixin Tan

Yu Yang

Qingfeng Lan

Jianfeng Lu

A. Rupam Mahmood

Pan Xu

2024-05-13

rl-conference.cc/RLC/2024/Conference (publié)

Code as Reward: Empowering Reinforcement Learning with VLMs

David Venuto

Sami Nur Islam

Martin Klissarov

Sherry Yang

Ankit Anand

Pre-trained Vision-Language Models (VLMs) are able to understand visual concepts, describe and decompose complex tasks into sub-tasks, and p… (voir plus)rovide feedback on task completion. In this paper, we aim to leverage these capabilities to support the training of reinforcement learning (RL) agents. In principle, VLMs are well suited for this purpose, as they can naturally analyze image-based observations and provide feedback (reward) on learning progress. However, inference in VLMs is computationally expensive, so querying them frequently to compute rewards would significantly slowdown the training of an RL agent. To address this challenge, we propose a framework named Code as Reward (VLM-CaR). VLM-CaR produces dense reward functions from VLMs through code generation, thereby significantly reducing the computational burden of querying the VLM directly. We show that the dense rewards generated through our approach are very accurate across a diverse set of discrete and continuous environments, and can be more effective in training RL policies than the original sparse environment rewards.

2024-04-30

ICML.cc/2024/Conference (spotlight)

Mixtures of Experts Unlock Parameter Scaling for Deep RL

Johan Obando-Ceron

Ghada Sokar

Timon Willi

Clare Lyle

Jesse Farebrother

Jakob Foerster

Karolina Dziugaite

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)

Nash Learning from Human Feedback

Remi Munos

Michal Valko

Daniele Calandriello

Mohammad Gheshlaghi Azar

Mark Rowland

Zhaohan Daniel Guo

Yunhao Tang

Matthieu Geist

Thomas Mesnard

Côme Fiegel

Andrea Michi

Marco Selvi

Sertan Girgin

Nikola Momchev

Olivier Bachem

Daniel J Mankowitz

Bilal Piot

Reinforcement learning from human feedback (RLHF) has emerged as the main paradigm for aligning large language models (LLMs) with human pref… (voir plus)erences. Traditionally, RLHF involves the initial step of learning a reward model from pairwise human feedback, i.e., expressed as preferences between pairs of text generations. Subsequently, the LLM’s policy is fine-tuned to maximize the reward through a reinforcement learning algorithm. In this study, we introduce an alternative pipeline for the fine-tuning of LLMs using pairwise human feedback. Our approach entails the initial learning of a pairwise preference model, which is conditioned on two inputs (instead of a single input in the case of a reward model) given a prompt, followed by the pursuit of a policy that consistently generates responses preferred over those generated by any competing policy, thus defining the Nash equilibrium of this preference model. We term this approach Nash learning from human feedback (NLHF). In the context of a tabular policy representation, we present a novel algorithmic solution, Nash-MD, founded on the principles of mirror descent. This algorithm produces a sequence of policies, with the last iteration converging to the regularized Nash equilibrium. Additionally, we explore parametric representations of policies and introduce gradient descent algorithms for deep-learning architectures. We illustrate the effectiveness of our approach by presenting experimental results on a text summarization task. We believe NLHF offers a compelling avenue for fine-tuning LLMs and enhancing the alignment of LLMs with human preferences.

2024-04-30

ICML.cc/2024/Conference (spotlight)

Discrete Probabilistic Inference as Control in Multi-path Environments

We consider the problem of sampling from a discrete and structured distribution as a sequential decision problem, where the objective is to … (voir plus)find a stochastic policy such that objects are sampled at the end of this sequential process proportionally to some predefined reward. While we could use maximum entropy Reinforcement Learning (MaxEnt RL) to solve this problem for some distributions, it has been shown that in general, the distribution over states induced by the optimal policy may be biased in cases where there are multiple ways to generate the same object. To address this issue, Generative Flow Networks (GFlowNets) learn a stochastic policy that samples objects proportionally to their reward by approximately enforcing a conservation of flows across the whole Markov Decision Process (MDP). In this paper, we extend recent methods correcting the reward in order to guarantee that the marginal distribution induced by the optimal MaxEnt RL policy is proportional to the original reward, regardless of the structure of the underlying MDP. We also prove that some flow-matching objectives found in the GFlowNet literature are in fact equivalent to well-established MaxEnt RL algorithms with a corrected reward. Finally, we study empirically the performance of multiple MaxEnt RL and GFlowNet algorithms on multiple problems involving sampling from discrete distributions.

2024-04-25

auai.org/UAI/2024/Conference (poster)