Continual Reinforcement Learning deployed in Real-life using Policy Distillation and Sim2Real Transfer

Abstract : We focus on the problem of teaching a robot to solve tasks presented sequentially, i.e., in a continual learning scenario. The robot should be able to solve all tasks it has encountered, without forgetting past tasks. We provide preliminary work on applying Reinforcement Learning to such setting, on 2D navigation tasks for a 3 wheel omni-directional robot. Our approach takes advantage of state representation learning and policy distillation. Policies are trained using learned features as input, rather than raw observations, allowing better sample efficiency. Policy distillation is used to combine multiple policies into a single one that solves all encountered tasks.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02285839
Contributor : Timothéee Lesort <>
Submitted on : Friday, September 13, 2019 - 10:37:22 AM
Last modification on : Wednesday, October 9, 2019 - 1:22:35 AM

Links full text

Identifiers

  • HAL Id : hal-02285839, version 1
  • ARXIV : 1906.04452

Citation

René Traoré, Hugo Caselles-Dupré, Timothée Lesort, Te Sun, Natalia Díaz-Rodríguez, et al.. Continual Reinforcement Learning deployed in Real-life using Policy Distillation and Sim2Real Transfer. ICML Workshop on “Multi-Task and Lifelong Reinforcement Learning”, Jun 2019, Long Beach, United States. ⟨hal-02285839⟩

Share

Metrics

Record views

27