Some Considerations on Learning to Explore via Meta-Reinforcement Learning

Bradly C. Stadie; Ge Yang; Rein Houthooft; Xi Chen; Yan Duan; Yuhuai Wu; Pieter Abbeel; Ilya Sutskever

Some Considerations on Learning to Explore via Meta-Reinforcement Learning

Artificial Intelligence 2019-01-15 v2

Authors: Bradly C. Stadie , Ge Yang , Rein Houthooft , Xi Chen , Yan Duan , Yuhuai Wu , Pieter Abbeel , Ilya Sutskever

Abstract

We consider the problem of exploration in meta reinforcement learning. Two new meta reinforcement learning algorithms are suggested: E-MAML and E- $\text{RL}^2$ . Results are presented on a novel environment we call `Krazy World' and a set of maze environments. We show E-MAML and E- $\text{RL}^2$ deliver better performance on tasks where exploration is important.

Keywords

reinforcement learning multi-view learning multi-agent reinforcement learning

Cite

@article{arxiv.1803.01118,
  title  = {Some Considerations on Learning to Explore via Meta-Reinforcement Learning},
  author = {Bradly C. Stadie and Ge Yang and Rein Houthooft and Xi Chen and Yan Duan and Yuhuai Wu and Pieter Abbeel and Ilya Sutskever},
  journal= {arXiv preprint arXiv:1803.01118},
  year   = {2019}
}

Related papers

View all related →

Artificial Intelligence · Computer Science

MetaCURE: Meta Reinforcement Learning with Empowerment-Driven Exploration