DC Field | Value | Language |
---|---|---|
dc.contributor.author | Han, Seungyul | ko |
dc.contributor.author | Sung, Youngchul | ko |
dc.date.accessioned | 2021-12-09T06:48:49Z | - |
dc.date.available | 2021-12-09T06:48:49Z | - |
dc.date.created | 2021-11-26 | - |
dc.date.created | 2021-11-26 | - |
dc.date.created | 2021-11-26 | - |
dc.date.issued | 2021-12-07 | - |
dc.identifier.citation | 35th Conference on Neural Information Processing Systems, NeurIPS 2021 | - |
dc.identifier.uri | http://hdl.handle.net/10203/290299 | - |
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.language | English | - |
dc.publisher | Neural Information Processing Systems | - |
dc.title | A max-min entropy framework for reinforcement learning | - |
dc.type | Conference | - |
dc.identifier.wosid | 000901616401049 | - |
dc.identifier.scopusid | 2-s2.0-85126629617 | - |
dc.type.rims | CONF | - |
dc.citation.publicationname | 35th Conference on Neural Information Processing Systems, NeurIPS 2021 | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Virtual | - |
dc.contributor.localauthor | Sung, Youngchul | - |
dc.contributor.nonIdAuthor | Han, Seungyul | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.