publications

google scholar → | dblp →

2023

A definition of continual reinforcement learning

David Abel, André Barreto, Benjamin Van Roy, Doina Precup, Hado Philip van Hasselt, Satinder Singh

Advances in Neural Information Processing Systems (NeurIPS)

[ pdf ] [ talk & poster ]

Deep reinforcement learning with plasticity injection

Evgenii Nikishin, Junhyuk Oh, Georg Ostrovski, Clare Lyle, Razvan Pascanu, Will Dabney, André Barreto

Advances in Neural Information Processing Systems (NeurIPS) - selected as a spotlight

[ pdf ] [ talk & poster ]

Simulation-based search

David Silver, André Barreto

International Congress of Mathematicians (ICM)

[ pdf ] [ talk ]

Temporal abstraction in reinforcement learning with the successor representation

Marlos C. Machado, André Barreto, Doina Precup, Michael Bowling

Journal of Machine Learning Research, v. 24 (1), pp. 3531–3599

[ pdf ]

On the convergence of bounded agents

David Abel, André Barreto, Hado van Hasselt, Benjamin Van Roy, Doina Precup, Satinder Singh

arXiv

[ pdf ]

2022

Generalised policy improvement with geometric policy composition

Shantanu Thakoor, Mark Rowland, Diana Borsa, Will Dabney, Rémi Munos, André Barreto

Proceedings of International Conference on Machine Learning (ICML) - selected as a spotlight

[ pdf ] [ talk & slides ] [ talk 2 ]

Approximate value equivalence

Christopher Grimm, André Barreto, Satinder Singh

Advances in Neural Information Processing Systems (NeurIPS)

[ pdf ] [ talk & slides ]

The phenomenon of policy churn

Tom Schaul, André Barreto, John Quan, Georg Ostrovski

Advances in Neural Information Processing Systems (NeurIPS)

[ pdf ] [ talk & slides ]

Model-value inconsistency as a signal for epistemic uncertainty

Angelos Filos, Eszter Vértes, Zita Marinho, Gregory Farquhar, Diana Borsa, Abram L. Friesen, Feryal Behbahani, Tom Schaul, André Barreto, Simon Osindero

Proceedings of International Conference on Machine Learning (ICML)

[ pdf ] [ talk & slides ]

Expressing non-Markov reward to a Markov agent

David Abel, André Barreto, Michael Bowling, Will Dabney, Steven Hansen, Anna Harutyunyan, Mark K. Ho, Ramana Kumar, Michael L. Littman, Doina Precup, Satinder Singh

The Multi-disciplinary Conference on Reinforcement Learning and Decision Making (RLDM)

[ pdf ]

Efficient information diffusion in time-varying graphs through deep reinforcement learning

Matheus R. F. Mendonça, André Barreto, Artur Ziviani

World Wide Web, volume 25, 2535–2560

[ more info ]

2021

Proper value equivalence

Christopher Grimm, André Barreto, Gregory Farquhar, David Silver, Satinder Singh

Advances in Neural Information Processing Systems (NeurIPS) - selected as a spotlight

[ pdf ] [ talk & slides ] [ talk ] [ more info]

Risk-aware transfer in reinforcement learning using successor features

Michael Gimelfarb, André Barreto, Scott Sanner, Chi-Guhn Lee

Advances in Neural Information Processing Systems (NeurIPS)

[ pdf ] [ talk & slides ] [ more info]

Discovering a set of policies for the worst case reward

Tom Zahavy, André Barreto, Daniel J Mankowitz, Shaobo Hou, Brendan O'Donoghue, Iurii Kemaev, Satinder Baveja Singh

International Conference on Learning Representations (ICLR) - selected as a spotlight

[ pdf ] [ talk & slides ] [ more info]

Temporally-extended ε-greedy exploration

Will Dabney, Georg Ostrovski, André Barreto

International Conference on Learning Representations (ICLR)

[ pdf ] [ talk ] [ more info ]

The value-improvement path: towards better representations for reinforcement learning

Will Dabney, André Barreto, Mark Rowland, Robert Dadashi, John Quan, Marc G. Bellemare, David Silver

AAAI Conference on Artificial Intelligence (AAAI)

[ pdf ] [ poster ]

Expected eligibility traces

Hado van Hasselt, Sephora Madjiheurem, Matteo Hessel, David Silver, André Barreto, Diana Borsa

AAAI Conference on Artificial Intelligence (AAAI) - selected as a distinguished paper

[ pdf ] [ poster ]

Discovering diverse nearly optimal policies with successor features

Tom Zahavy, Brendan O'Donoghue, André Barreto, Volodymyr Mnih, Sebastian Flennerhag, Satinder Singh

arXiv

[ pdf ]

Beyond fine-tuning: transferring behavior in reinforcement learning

Víctor Campos, Pablo Sprechmann, Steven Hansen, André Barreto, Steven Kapturowski, Alex Vitvitskyi, Adrià Puigdomènech Badia, Charles Blundell

arXiv

[ pdf ]

New machine learning and physics-based scoring functions for drug discovery

Isabella Guedes, André Barreto, Diogo Marinho, Eduardo Krempser, Mélaine Kuenemann, Olivier Sperandio, Laurent Dardenne, Maria A. Miteva

Scientific Reports

[ more info]

2020

Fast reinforcement learning with generalized policy updates

André Barreto, Shaobo Hou, Diana Borsa, David Silver, Doina Precup

Proceedings of the National Academy of Sciences

[ pdf ] [ blog post] [ talk ] [ talk 2 ] [ more info ]

The value equivalence principle for model-based reinforcement learning

Christopher Grimm, André Barreto, Satinder Singh, David Silver

Advances in Neural Information Processing Systems (NeurIPS)

[ pdf ] [ slides ] [ talk ]

On efficiency in hierarchical reinforcement learning

Zheng Wen, Doina Precup, Morteza Ibrahimi, André Barreto, Benjamin Van Roy, Satinder Singh

Advances in Neural Information Processing Systems (NeurIPS)

[ pdf ] [ talk ]

Fast task inference with variational intrinsic successor features

Steven Hansen, Will Dabney, André Barreto, David Warde-Farley, Tom Van de Wiele, Volodymyr Mnih

Proceedings of the International Conference on Learning Representations (ICLR)

[ pdf ] [ talk] [ more info ]

Approximating network centrality measures using node embedding and machine learning

Matheus Mendonça, André Barreto, Artur Ziviani

IEEE Transactions on Network Science and Engineering

[ more info ]

Efficient information diffusion in time-varying graphs through deep reinforcement learning

Matheus Mendonça, André Barreto, Artur Ziviani

arXiv

[ pdf ]

Temporal difference uncertainties as a signal for exploration

Sebastian Flennerhag, Jane X. Wang, Pablo Sprechmann, Francesco Visin, Alexandre Galashov, Steven Kapturowski, Diana L. Borsa, Nicolas Heess, André Barreto, Razvan Pascanu

arXiv

[ pdf ]

2019

The option keyboard: combining skills in reinforcement learning

André Barreto, Diana Borsa, Shaobo Hou, Gheorghe Comanici, Eser Aygün, Philippe Hamel, Daniel Toyama, Jonathan J. Hunt, Shibl Mourad, David Silver, Doina Precup

Advances in Neural Information Processing Systems (NeurIPS)

[ pdf ] [ video ] [ slides ] [ poster ] [ more info ]

Universal successor features approximators

Diana Borsa, André Barreto, John Quan, Daniel Mankowitz, Rémi Munos, Hado van Hasselt, David Silver, Tom Schaul

Proceedings of the International Conference on Learning Representations (ICLR)

[ pdf ] [ video 1 ] [ video 2 ] [ more info ]

Composing entropic policies using divergence correction

Jonathan Hunt, André Barreto, Timothy Lillicrap, Nicolas Heess

Proceedings of International Conference on Machine Learning (ICML)

[ pdf ] [ videos ] [ more info ]

Adaptive temporal-difference learning for policy evaluation with per-state uncertainty estimates

Carlos Riquelme, Hugo Penedones, Damien Vincent, Hartmut Maennel, Timothy A. Mann, André Barreto, Sylvain Gelly, Gergely Neu

Advances in Neural Information Processing Systems (NeurIPS)

[ pdf ] [ poster ] [ more info ]

General non-linear Bellman equations

Hado van Hasselt, John Quan, Matteo Hessel, Zhongwen Xu, Diana Borsa, André Barreto

The Multi-disciplinary Conference on Reinforcement Learning and Decision Making (RLDM)

publications

2023

A definition of continual reinforcement learning

Deep reinforcement learning with plasticity injection

Simulation-based search

Temporal abstraction in reinforcement learning with the successor representation

On the convergence of bounded agents

2022

Generalised policy improvement with geometric policy composition

Approximate value equivalence

The phenomenon of policy churn

Model-value inconsistency as a signal for epistemic uncertainty

Expressing non-Markov reward to a Markov agent

Efficient information diffusion in time-varying graphs through deep reinforcement learning

2021

Proper value equivalence

Risk-aware transfer in reinforcement learning using successor features

Discovering a set of policies for the worst case reward

Temporally-extended ε-greedy exploration

The value-improvement path: towards better representations for reinforcement learning

Expected eligibility traces

Discovering diverse nearly optimal policies with successor features

Beyond fine-tuning: transferring behavior in reinforcement learning

New machine learning and physics-based scoring functions for drug discovery

2020

Fast reinforcement learning with generalized policy updates

The value equivalence principle for model-based reinforcement learning

On efficiency in hierarchical reinforcement learning

Fast task inference with variational intrinsic successor features

Approximating network centrality measures using node embedding and machine learning

Efficient information diffusion in time-varying graphs through deep reinforcement learning

Temporal difference uncertainties as a signal for exploration

2019

The option keyboard: combining skills in reinforcement learning

Universal successor features approximators

Composing entropic policies using divergence correction

Adaptive temporal-difference learning for policy evaluation with per-state uncertainty estimates

General non-linear Bellman equations

Graph-based skill acquisition for reinforcement learning

2018

Transfer in deep reinforcement learning using successor features and generalised policy improvement

Fast deep reinforcement learning using online adjustments from the past

Online TD(λ) for discrete-time Markov jump linear systems

Temporal difference learning with neural networks - study of the leakage propagation problem

Unicorn: continual learning with a universal, off-Policy, agent

Abstract state transition graphs for model-based reinforcement learning

2017

Successor features for transfer in reinforcement learning

The Predictron: end-to-end learning and planning

Natural value approximators: learning when to trust past estimates

Value-aware loss function for model-based reinforcement learning

Count-based quadratic control of Markov jump linear systems with unknown transition probabilities

Transfer on count-based quadratic control of Markov jump linear systems with unknown transition probabilities

2016

Practical kernel-based reinforcement learning

Incremental stochastic factorization for online reinforcement learning

2015

An expectation-maximization algorithm to compute a stochastic factorization from data

Classification-based approximate policy iteration

2014

Policy iteration based on stochastic factorization

Tree-based on-line reinforcement learning

Starting to uncover the relationship between stochastic factorization and hidden Markov models

2013

CAPI: Generalized classification-based approximate policy iteration

Reinforcement learning competition 2013: Controllability and kernel-­based stochastic factorization

2012

On-line reinforcement learning using incremental kernel-based stochastic factorization

Analysis of composition-based metagenomic classification

2011

Computing the stationary distribution of a finite Markov chain through stochastic factorization

Reinforcement learning using kernel-based stochastic factorization

Lumping the states of a finite Markov chain through stochastic factorization

Exploring performance proﬁles for analyzing benchmark experiments

Evolving numerical constants in grammatical evolution with the ephemeral constant method

A new approach for generating numerical constants in grammatical evolution

...2010

Probabilistic performance proﬁles for the experimental evaluation of stochastic algorithms

Kernel-based stochastic factorization for batch reinforcement learning

On the characteristics of sequential decision problems and their impact on evolutionary computation and reinforcement learning

Reinforcement learning competition 2013: Controllability and kernel-based stochastic factorization