DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kim, Kee-Eung | - |
dc.contributor.advisor | 김기응 | - |
dc.contributor.author | Lee, Byung-Jun | - |
dc.date.accessioned | 2022-04-21T19:34:23Z | - |
dc.date.available | 2022-04-21T19:34:23Z | - |
dc.date.issued | 2021 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=956453&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/295721 | - |
dc.description | 학위논문(박사) - 한국과학기술원 : 전산학부, 2021.2,[iv, 61 p. :] | - |
dc.description.abstract | Offline reinforcement learning (RL) aims to learn without additional interaction with the environment, based on the pre-collected dataset. It has recently gathered attention due to its promise for real-world applications. Unlike online RL where the agent's predictions can be further corrected through additional interactions, offline RL requires robust policy improvement under the potentially incorrect predictions. To do this, it is necessary to accurately measure the uncertainty of the implicitly or explicitly constructed environment model, and design an algorithm that can find a trade-off between the potential policy performance and the uncertainty in policy evaluation. In this thesis, we study offline RL algorithms about (1) finding a good trade-off using a validation split and (2) learning model that is more robust especially for offline RL. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Machine Learning▼aReinforcement Learning▼aOffline Reinforcement Learning▼aHypergradient▼aBalanced Representation | - |
dc.subject | 기계학습▼a강화학습▼a오프라인 강화학습▼a하이퍼그래디언트▼a표현 밸런싱 | - |
dc.title | Algorithms for efficient offline reinforcement learning | - |
dc.title.alternative | 효율적인 오프라인 강화학습을 위한 알고리즘 연구 | - |
dc.type | Thesis(Ph.D) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전산학부, | - |
dc.contributor.alternativeauthor | 이병준 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.