DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Chung, Sae-Young | - |
dc.contributor.advisor | 정세영 | - |
dc.contributor.author | Lee, Su Young | - |
dc.date.accessioned | 2019-09-04T02:44:15Z | - |
dc.date.available | 2019-09-04T02:44:15Z | - |
dc.date.issued | 2019 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=843413&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/266919 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2019.2,[iii, 29 p. :] | - |
dc.description.abstract | We propose Episodic Backward Update – a new algorithm to boost the performance of a deep reinforcement learning agent by a fast reward propagation. In contrast to the conventional use of the experience replay with uniform random sampling, our agent samples a whole episode and successively propagates the value of a state to its previous states. Our computationally efficient recursive algorithm allows sparse and delayed rewards to propagate efficiently through all transitions of a sampled episode. We evaluate our algorithm on 2D MNIST maze environment and 49 games of the Atari 2600 environment, and show that our method improves sample efficiency with a competitive amount of computational cost. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | deep reinforcement learning▼adeep Q-learning▼adeep neural network▼aexperience replay▼asample efficiency | - |
dc.subject | 심층강화학습▼a심층 Q 러닝▼a심층 인공 신경망▼a경험 재현▼a샘플 효율성 | - |
dc.title | Sample-efficient deep reinforcement learning via episodic backward update | - |
dc.title.alternative | 에피소드 후향 업데이트를 통한 효율적인 심층강화학습 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전기및전자공학부, | - |
dc.contributor.alternativeauthor | 이수영 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.