Timezone: »
Efficient exploration remains a major challenge for reinforcement learning (RL). Common dithering strategies for exploration, such as epsilon-greedy, do not carry out temporally-extended (or deep) exploration; this can lead to exponentially larger data requirements. However, most algorithms for statistically efficient RL are not computationally tractable in complex environments. Randomized value functions offer a promising approach to efficient exploration with generalization, but existing algorithms are not compatible with nonlinearly parameterized value functions. As a first step towards addressing such contexts we develop bootstrapped DQN. We demonstrate that bootstrapped DQN can combine deep exploration with deep neural networks for exponentially faster learning than any dithering strategy. In the Arcade Learning Environment bootstrapped DQN substantially improves learning speed and cumulative performance across most games.
Author Information
Ian Osband (DeepMind)
Charles Blundell (DeepMind)
Alexander Pritzel (Google Deepmind)
Benjamin Van Roy (Stanford University)
More from the Same Authors
-
2020 Poster: Pointer Graph Networks »
Petar Veličković · Lars Buesing · Matthew Overlan · Razvan Pascanu · Oriol Vinyals · Charles Blundell -
2020 Spotlight: Pointer Graph Networks »
Petar Veličković · Lars Buesing · Matthew Overlan · Razvan Pascanu · Oriol Vinyals · Charles Blundell -
2019 Poster: Generalization of Reinforcement Learners with Working and Episodic Memory »
Meire Fortunato · Melissa Tan · Ryan Faulkner · Steven Hansen · Adrià Puigdomènech Badia · Gavin Buttimore · Charles Deck · Joel Leibo · Charles Blundell -
2019 Poster: Information-Theoretic Confidence Bounds for Reinforcement Learning »
Xiuyuan Lu · Benjamin Van Roy -
2018 Poster: An Information-Theoretic Analysis for Thompson Sampling with Many Actions »
Shi Dong · Benjamin Van Roy -
2018 Poster: Scalable Coordinated Exploration in Concurrent Reinforcement Learning »
Maria Dimakopoulou · Ian Osband · Benjamin Van Roy -
2018 Poster: Randomized Prior Functions for Deep Reinforcement Learning »
Ian Osband · John Aslanides · Albin Cassirer -
2018 Spotlight: Randomized Prior Functions for Deep Reinforcement Learning »
Ian Osband · John Aslanides · Albin Cassirer -
2018 Poster: Fast deep reinforcement learning using online adjustments from the past »
Steven Hansen · Alexander Pritzel · Pablo Sprechmann · Andre Barreto · Charles Blundell -
2017 Poster: Ensemble Sampling »
Xiuyuan Lu · Benjamin Van Roy -
2017 Poster: Conservative Contextual Linear Bandits »
Abbas Kazerouni · Mohammad Ghavamzadeh · Yasin Abbasi · Benjamin Van Roy -
2017 Poster: Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles »
Balaji Lakshminarayanan · Alexander Pritzel · Charles Blundell -
2017 Spotlight: Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles »
Balaji Lakshminarayanan · Alexander Pritzel · Charles Blundell -
2016 Poster: Matching Networks for One Shot Learning »
Oriol Vinyals · Charles Blundell · Timothy Lillicrap · koray kavukcuoglu · Daan Wierstra -
2014 Workshop: Advances in Variational Inference »
David Blei · Shakir Mohamed · Michael Jordan · Charles Blundell · Tamara Broderick · Matthew D. Hoffman -
2014 Workshop: Large-scale reinforcement learning and Markov decision problems »
Benjamin Van Roy · Mohammad Ghavamzadeh · Peter Bartlett · Yasin Abbasi Yadkori · Ambuj Tewari -
2014 Poster: Near-optimal Reinforcement Learning in Factored MDPs »
Ian Osband · Benjamin Van Roy -
2014 Poster: Learning to Optimize via Information-Directed Sampling »
Daniel Russo · Benjamin Van Roy -
2014 Spotlight: Near-optimal Reinforcement Learning in Factored MDPs »
Ian Osband · Benjamin Van Roy -
2014 Poster: Model-based Reinforcement Learning and the Eluder Dimension »
Ian Osband · Benjamin Van Roy -
2013 Poster: (More) Efficient Reinforcement Learning via Posterior Sampling »
Ian Osband · Daniel Russo · Benjamin Van Roy -
2013 Poster: Eluder Dimension and the Sample Complexity of Optimistic Exploration »
Daniel Russo · Benjamin Van Roy -
2013 Oral: Eluder Dimension and the Sample Complexity of Optimistic Exploration »
Daniel Russo · Benjamin Van Roy -
2013 Poster: Bayesian Hierarchical Community Discovery »
Charles Blundell · Yee Whye Teh -
2013 Poster: Efficient Exploration and Value Function Generalization in Deterministic Systems »
Zheng Wen · Benjamin Van Roy -
2012 Poster: Modelling Reciprocating Relationships with Hawkes processes »
Charles Blundell · Katherine Heller · Jeff Beck -
2012 Spotlight: Modelling Reciprocating Relationships with Hawkes processes »
Charles Blundell · Katherine Heller · Jeff Beck -
2012 Poster: Efficient Reinforcement Learning for High Dimensional Linear Quadratic Systems »
Morteza Ibrahimi · Adel Javanmard · Benjamin Van Roy -
2011 Poster: Modelling Genetic Variations using Fragmentation-Coagulation Processes »
Yee Whye Teh · Charles Blundell · Lloyd T Elliott -
2011 Oral: Modelling Genetic Variations using Fragmentation-Coagulation Processes »
Yee Whye Teh · Charles Blundell · Lloyd T Elliott -
2009 Poster: Directed Regression »
Yi-Hao Kao · Benjamin Van Roy · Xiang Yan