HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Conference papers

Self-Imitation Advantage Learning

Johan Ferret 1, 2 Olivier Pietquin 1 Matthieu Geist 1
2 Scool - Scool
Inria Lille - Nord Europe, CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189
Abstract : Self-imitation learning is a Reinforcement Learning (RL) method that encourages actions whose returns were higher than expected, which helps in hard exploration and sparse reward problems. It was shown to improve the performance of on-policy actor-critic methods in several discrete control tasks. Nevertheless, applying self-imitation to the mostly action-value based off-policy RL methods is not straightforward. We propose SAIL, a novel generalization of self-imitation learning for off-policy RL, based on a modification of the Bellman optimality operator that we connect to Advantage Learning. Crucially, our method mitigates the problem of stale returns by choosing the most optimistic return estimate between the observed return and the current action-value for self-imitation. We demonstrate the empirical effectiveness of SAIL on the Arcade Learning Environment, with a focus on hard exploration games.
Document type :
Conference papers
Complete list of metadata

Contributor : Johan Ferret Connect in order to contact the contributor
Submitted on : Thursday, March 4, 2021 - 5:13:53 PM
Last modification on : Thursday, March 24, 2022 - 3:42:45 AM
Long-term archiving on: : Saturday, June 5, 2021 - 7:17:38 PM


Files produced by the author(s)


  • HAL Id : hal-03159815, version 1


Johan Ferret, Olivier Pietquin, Matthieu Geist. Self-Imitation Advantage Learning. AAMAS 2021 - 20th International Conference on Autonomous Agents and Multiagent Systems, May 2021, Londres / Virtual, United Kingdom. ⟨hal-03159815⟩



Record views


Files downloads