Skip to Main content Skip to Navigation

Fast active learning for pure exploration in reinforcement learning

Pierre Ménard 1 Omar Domingues 1 Anders Jonsson 2 Emilie Kaufmann Edouard Leurent 3 Michal Valko 1, 4
1 SEQUEL - Sequential Learning
Inria Lille - Nord Europe, CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189
Abstract : Realistic environments often provide agents with very limited feedback. When the environment is initially unknown, the feedback, in the beginning, can be completely absent, and the agents may first choose to devote all their effort on exploring efficiently. The exploration remains a challenge while it has been addressed with many hand-tuned heuristics with different levels of generality on one side, and a few theoretically backed exploration strategies on the other. Many of them are incarnated by intrinsic motivation and in particular explorations bonuses. A common rule of thumb for exploration bonuses is to use 1/ √ n bonus that is added to the empirical estimates of the reward, where n is a number of times this particular state (or a state-action pair) was visited. We show that, surprisingly, for a pure-exploration objective of reward-free exploration, bonuses that scale with 1/n bring faster learning rates, improving the known upper bounds with respect to the dependence on the horizon H. Furthermore, we show that with an improved analysis of the stopping time, we can improve by a factor H the sample complexity in the best-policy identification setting, which is another pure-exploration objective, where the environment provides rewards but the agent is not penalized for its behavior during the exploration phase.
Document type :
Complete list of metadata

Cited literature [3 references]  Display  Hide  Download
Contributor : Michal Valko Connect in order to contact the contributor
Submitted on : Sunday, July 26, 2020 - 11:53:24 PM
Last modification on : Friday, December 17, 2021 - 11:28:34 AM
Long-term archiving on: : Tuesday, December 1, 2020 - 5:51:38 PM


Files produced by the author(s)


  • HAL Id : hal-02906985, version 1


Pierre Ménard, Omar Domingues, Anders Jonsson, Emilie Kaufmann, Edouard Leurent, et al.. Fast active learning for pure exploration in reinforcement learning. [Research Report] DeepMind. 2020. ⟨hal-02906985v1⟩



Les métriques sont temporairement indisponibles