dc.citation.conferencePlace |
US |
- |
dc.citation.conferencePlace |
Virtual-only |
- |
dc.citation.title |
Neural Information Processing Systems |
- |
dc.contributor.author |
Han, Seungyul |
- |
dc.contributor.author |
Sung, Youngchul |
- |
dc.date.accessioned |
2024-01-31T21:06:24Z |
- |
dc.date.available |
2024-01-31T21:06:24Z |
- |
dc.date.created |
2021-11-28 |
- |
dc.date.issued |
2021-12-08 |
- |
dc.description.abstract |
In this paper, we propose a max-min entropy framework for reinforcement learning (RL) to overcome the limitation of the soft actor-critic (SAC) algorithm implementing the maximum entropy RL in model-free sample-based learning. Whereas the maximum entropy RL guides learning for policies to reach states with high entropy in the future, the proposed max-min entropy framework aims to learn to visit states with low entropy and maximize the entropy of these low-entropy states to promote better exploration. For general Markov decision processes (MDPs), an efficient algorithm is constructed under the proposed max-min entropy framework based on disentanglement of exploration and exploitation. Numerical results show that the proposed algorithm yields drastic performance improvement over the current state-of-the-art RL algorithms. |
- |
dc.identifier.bibliographicCitation |
Neural Information Processing Systems |
- |
dc.identifier.scopusid |
2-s2.0-85126629617 |
- |
dc.identifier.uri |
https://scholarworks.unist.ac.kr/handle/201301/76467 |
- |
dc.identifier.url |
https://papers.nips.cc/paper/2021/hash/d7b76edf790923bf7177f7ebba5978df-Abstract.html |
- |
dc.publisher |
Neural Information Processing Systems |
- |
dc.title |
A Max-Min Entropy Framework for Reinforcement Learning |
- |
dc.type |
Conference Paper |
- |
dc.date.conferenceDate |
2021-12-06 |
- |