DisCoRL: Continual Reinforcement Learning via Policy Distillation
Abstract
In multi-task reinforcement learning there are two main challenges: at training time, the ability to learn different policies with a single model; at test time, inferring which of those policies applying without an external signal. In the case of continual reinforcement learning a third challenge arises: learning tasks sequentially without forgetting the previous ones. In this paper, we tackle these challenges by proposing DisCoRL, an approach combining state representation learning and policy distillation. We experiment on a sequence of three simulated 2D navigation tasks with a 3 wheel omni-directional robot. Moreover, we tested our approach's robustness by transferring the final policy into a real life setting. The policy can solve all tasks and automatically infer which one to run.
- Publication:
-
arXiv e-prints
- Pub Date:
- July 2019
- DOI:
- 10.48550/arXiv.1907.05855
- arXiv:
- arXiv:1907.05855
- Bibcode:
- 2019arXiv190705855T
- Keywords:
-
- Computer Science - Machine Learning;
- Computer Science - Artificial Intelligence;
- Statistics - Machine Learning
- E-Print:
- arXiv admin note: text overlap with arXiv:1906.04452