Doina Precup

Sumana Basu

PhD - McGill University

Co-supervisor :

Adriana Romero Soriano

PhD - McGill University

Lynn Cherif

Master's Research - McGill University

Co-supervisor :

PhD - McGill University

Co-supervisor :

PhD - McGill University

Principal supervisor :

David Meger

Jonathan Colaço Carr

Master's Research - McGill University

Principal supervisor :

Prakash Panangaden

Élodie Coté-Gauthier

Collaborating researcher - McGill University

Franco Del Balso

Research Intern - Université de Montréal

Jesse Farebrother

PhD - McGill University

Principal supervisor :

Marc Gendron-Bellemare

PhD - McGill University

Principal supervisor :

Eilif Benjamin Muller

PhD - McGill University

Haque Ishfaq

PhD - McGill University

Mohammad Sami Nur Islam Islam

Master's Research - McGill University

Arushi Jain

PhD - McGill University

PhD - McGill University

Postdoctorate - McGill University

Elaine Lau

Master's Research - McGill University

Jonathan Lebensold

Collaborating Alumni - McGill University

Undergraduate - McGill University

Ray Luo

PhD - McGill University

Principal supervisor :

G McCracken

PhD - McGill University

Nazanin Mohammadi Sepahvand

PhD - McGill University

Shahrad Mohammadzadeh

Master's Research - McGill University

Principal supervisor :

Gabriela Moisescu-Pareja

Collaborating researcher - McGill University

Co-supervisor :

Irina Rish

Padideh Nouri

PhD - Université de Montréal

Co-supervisor :

Charles Onu

PhD - McGill University

PhD - McGill University

Co-supervisor :

Nate Rahn

PhD - McGill University

Principal supervisor :

Marc Gendron-Bellemare

Sahand Rezaei-Shoshtari

PhD - McGill University

Co-supervisor :

PhD - McGill University

Co-supervisor :

PhD - McGill University

Co-supervisor :

Blake Richards

samiemandana@gmail.com

PhD - McGill University

Nishanth Anand Vemgal

PhD - McGill University

PhD - McGill University

Priyesh Vijayan

PhD - McGill University

Co-supervisor :

Samira Ebrahimi Kahou

Zihan Wang

PhD - McGill University

Guangyuan Wang

Research Intern - McGill University

Steve Wen

Master's Research - McGill University

Co-supervisor :

Gregory Dudek

Zijing Wu

PhD - McGill University

Co-supervisor :

PhD - McGill University

Skipper: Combining Spatial and Temporal Abstraction for Better Generalization

Harry Zhao

Collaborating Alumni - McGill University

Co-supervisor :

Blog Posts

Generic thumbnail for Mila Blog articles.

February 22, 2024

Mingde Harry Zhao

Safa Alver

Harm van Seijen

Romain Laroche

Doina Precup

Yoshua Bengio

Read the article

Publications

What can I do here? A Theory of Affordances in Reinforcement Learning

Khimya Khetarpal

Zafarali Ahmed

Gheorghe Comanici

David Abel

2020-11-21

Proceedings of the 37th International Conference on Machine Learning (published)

proceedings.mlr.press

Gradient Starvation: A Learning Proclivity in Neural Networks

Sékou-Oumar Kaba

We identify and formalize a fundamental gradient descent phenomenon resulting in a learning proclivity in over-parameterized neural networks… (see more). Gradient Starvation arises when cross-entropy loss is minimized by capturing only a subset of features relevant for the task, despite the presence of other predictive features that fail to be discovered. This work provides a theoretical explanation for the emergence of such feature imbalance in neural networks. Using tools from Dynamical Systems theory, we identify simple properties of learning dynamics during gradient descent that lead to this imbalance, and prove that such a situation can be expected given certain statistical structure in training data. Based on our proposed formalism, we develop guarantees for a novel regularization method aimed at decoupling feature learning dynamics, improving accuracy and robustness in cases hindered by gradient starvation. We illustrate our findings with simple and real-world out-of-distribution (OOD) generalization experiments.

2020-11-18

ArXiv (preprint)

Diversity-Enriched Option-Critic

Anand Kamat

Temporal abstraction allows reinforcement learning agents to represent knowledge and develop strategies over different temporal scales. The … (see more)option-critic framework has been demonstrated to learn temporally extended actions, represented as options, end-to-end in a model-free setting. However, feasibility of option-critic remains limited due to two major challenges, multiple options adopting very similar behavior, or a shrinking set of task relevant options. These occurrences not only void the need for temporal abstraction, they also affect performance. In this paper, we tackle these problems by learning a diverse set of options. We introduce an information-theoretic intrinsic reward, which augments the task reward, as well as a novel termination objective, in order to encourage behavioral diversity in the option set. We show empirically that our proposed method is capable of learning options end-to-end on several discrete and continuous control tasks, outperforms option-critic by a wide margin. Furthermore, we show that our approach sustainably generates robust, reusable, reliable and interpretable options, in contrast to option-critic.

2020-11-04

ArXiv (preprint)

A Study of Policy Gradient on a Class of Exactly Solvable Models

Gavin McCracken

Colin Daniels

Rosie Zhao

Anna M. Brandenberger

Prakash Panangaden

Policy gradient methods are extensively used in reinforcement learning as a way to optimize expected return. In this paper, we explore the e… (see more)volution of the policy parameters, for a special class of exactly solvable POMDPs, as a continuous-state Markov chain, whose transition probabilities are determined by the gradient of the distribution of the policy's value. Our approach relies heavily on random walk theory, specifically on affine Weyl groups. We construct a class of novel partially observable environments with controllable exploration difficulty, in which the value distribution, and hence the policy parameter evolution, can be derived analytically. Using these environments, we analyze the probabilistic convergence of policy gradient to different local maxima of the value function. To our knowledge, this is the first approach developed to analytically compute the landscape of policy gradient in POMDPs for a class of such environments, leading to interesting insights into the difficulty of this problem.

2020-11-03

ArXiv (preprint)

A Fully Tensorized Recurrent Neural Network

Charles Onu

Jacob Miller

2020-10-08

ArXiv (preprint)

Keynote Lecture - Building Knowledge For AI AgentsWith Reinforcement Learning

Summary form only given, as follows. The complete presentation was not made available for publication as part of the conference proceedings.… (see more) Reinforcement learning allows autonomous agents to learn how to act in a stochastic, unknown environment, with which they can interact. Deep reinforcement learning, in particular, has achieved great success in well-defined application domains, such as Go or chess, in which an agent has to learn how to act and there is a clear success criterion. In this talk, I will focus on the potential role of reinforcement learning as a tool for building knowledge representations in AI agents whose goal is to perform continual learning. I will examine a key concept in reinforcement learning, the value function, and discuss its generalization to support various forms of predictive knowledge. I will also discuss the role of temporally extended actions, and their associated predictive models, in learning procedural knowledge. In order to tame the possible complexity of learning knowledge representations, reinforcement learning agents can use the concepts of intents (ie intended consequences of courses of actions) and affordances (which capture knowlege about where actions can be applied). Finally, I will discuss the challenge of how to evaluate reinforcement learning agents whose goal is not just to control their environment, but also to build knowledge about their world.

2020-09-03

International Conference on Computational Photography (published)

doi.org

Fast reinforcement learning with generalized policy updates

Andre Barreto

Shaobo Hou

Diana Borsa

David Silver

The combination of reinforcement learning with deep learning is a promising approach to tackle important sequential decision-making problems… (see more) that are currently intractable. One obstacle to overcome is the amount of data needed by learning systems of this type. In this article, we propose to address this issue through a divide-and-conquer approach. We argue that complex decision problems can be naturally decomposed into multiple tasks that unfold in sequence or in parallel. By associating each task with a reward function, this problem decomposition can be seamlessly accommodated within the standard reinforcement-learning formalism. The specific way we do so is through a generalization of two fundamental operations in reinforcement learning: policy improvement and policy evaluation. The generalized version of these operations allow one to leverage the solution of some tasks to speed up the solution of others. If the reward function of a task can be well approximated as a linear combination of the reward functions of tasks previously solved, we can reduce a reinforcement-learning problem to a simpler linear regression. When this is not the case, the agent can still exploit the task solutions by using them to interact with and learn about the environment. Both strategies considerably reduce the amount of data needed to solve a reinforcement-learning problem.

2020-08-17

Proceedings of the National Academy of Sciences of the United States of America (published)

doi.org

A Brief Look at Generalization in Visual Meta-Reinforcement Learning

Safa Alver

Due to the realization that deep reinforcement learning algorithms trained on high-dimensional tasks can strongly overfit to their training … (see more)environments, there have been several studies that investigated the generalization performance of these algorithms. However, there has been no similar study that evaluated the generalization performance of algorithms that were specifically designed for generalization, i.e. meta-reinforcement learning algorithms. In this paper, we assess the generalization performance of these algorithms by leveraging high-dimensional, procedurally generated environments. We find that these algorithms can display strong overfitting when they are evaluated on challenging tasks. We also observe that scalability to high-dimensional tasks with sparse rewards remains a significant problem among many of the current meta-reinforcement learning algorithms. With these results, we highlight the need for developing meta-reinforcement learning algorithms that can both generalize and scale.

2020-07-13

ICML.cc/2020/Workshop/LifelongML (unknown)

openreview.net

Learning to Prove from Synthetic Theorems

Eser Aygün

Zafarali Ahmed

Ankit Anand

Vlad Firoiu

Xavier Glorot

Laurent Orseau

Shibl Mourad

A major challenge in applying machine learning to automated theorem proving is the scarcity of training data, which is a key ingredient in t… (see more)raining successful deep learning models. To tackle this problem, we propose an approach that relies on training with synthetic theorems, generated from a set of axioms. We show that such theorems can be used to train an automated prover and that the learned prover transfers successfully to human-generated theorems. We demonstrate that a prover trained exclusively on synthetic theorems can solve a substantial fraction of problems in TPTP, a benchmark dataset that is used to compare state-of-the-art heuristic provers. Our approach outperforms a model trained on human-generated problems in most axiom sets, thereby showing the promise of using synthetic data for this task.

2020-06-19

ArXiv (preprint)

Efficient Planning under Partial Observability with Unnormalized Q Functions and Spectral Learning

Tianyu Li

Bogdan Mazoure

Guillaume Rabusseau

2020-06-03

Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics (published)

proceedings.mlr.press

Value Preserving State-Action Abstractions

David Abel

Nathan Umbanhowar

Khimya Khetarpal

Dilip Arumugam

Michael L. Littman

Abstraction can improve the sample efficiency of reinforcement learning. However, the process of abstraction inherently discards information… (see more), potentially compromising an agent’s ability to represent high-value policies. To mitigate this, we here introduce combinations of state abstractions and options that are guaranteed to preserve the representation of near-optimal policies. We first define φ-relative options, a general formalism for analyzing the value loss of options paired with a state abstraction, and present necessary and sufficient conditions for φ-relative options to preserve near-optimal behavior in any finite Markov Decision Process. We further show that, under appropriate assumptions, φ-relative options can be composed to induce hierarchical abstractions that are also guaranteed to represent high-value policies.ion can improve the sample efficiency of reinforcement learning. However, the process of abstraction inherently discards information, potentially compromising an agent’s ability to represent high-value policies. To mitigate this, we here introduce combinations of state abstractions and options that are guaranteed to preserve the representation of near-optimal policies. We first define φ-relative options, a general formalism for analyzing the value loss of options paired with a state abstraction, and present necessary and sufficient conditions for φ-relative options to preserve near-optimal behavior in any finite Markov Decision Process. We further show that, under appropriate assumptions, φ-relative options can be composed to induce hierarchical abstractions that are also guaranteed to represent high-value policies.

2020-06-03

International Conference on Artificial Intelligence and Statistics (published)

proceedings.mlr.press

Value Preserving State-Action Abstractions

David Abel

Nathan Umbanhowar

Khimya Khetarpal

Dilip Arumugam