No-regret shannon entropy regularized neural contextual bandit online learning for robotic grasping
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Lee, K. | - |
dc.contributor.author | Choy, J. | - |
dc.contributor.author | Choi, Y. | - |
dc.contributor.author | Kee, H. | - |
dc.contributor.author | Oh, S. | - |
dc.date.accessioned | 2022-11-28T01:57:51Z | - |
dc.date.available | 2022-11-28T01:57:51Z | - |
dc.date.issued | 2020-10 | - |
dc.identifier.issn | 2153-0858 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/59359 | - |
dc.description.abstract | In this paper, we propose a novel contextual bandit algorithm that employs a neural network as a reward estimator and utilizes Shannon entropy regularization to encourage exploration, which is called Shannon entropy regularized neural contextual bandits (SERN). In many learning-based algorithms for robotic grasping, the lack of the real-world data hampers the generalization performance of a model and makes it difficult to apply a trained model to real-world problems. To handle this issue, the proposed method utilizes the benefit of an online learning. The proposed method trains a neural network to predict the success probability of a given grasp pose based on a depth image, which is called a grasp quality. We theoretically show that the SERN has a no regret property. We empirically demonstrate that the SERN outperforms ϵ-greedy in terms of sample efficiency. | - |
dc.format.extent | 6 | - |
dc.language | 영어 | - |
dc.language.iso | ENG | - |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | - |
dc.title | No-regret shannon entropy regularized neural contextual bandit online learning for robotic grasping | - |
dc.type | Article | - |
dc.identifier.doi | 10.1109/IROS45743.2020.9341123 | - |
dc.identifier.bibliographicCitation | IEEE International Conference on Intelligent Robots and Systems, pp 9620 - 9625 | - |
dc.description.isOpenAccess | N | - |
dc.identifier.scopusid | 2-s2.0-85102409713 | - |
dc.citation.endPage | 9625 | - |
dc.citation.startPage | 9620 | - |
dc.citation.title | IEEE International Conference on Intelligent Robots and Systems | - |
dc.type.docType | Conference Paper | - |
dc.publisher.location | 미국 | - |
dc.subject.keywordPlus | Agricultural robots | - |
dc.subject.keywordPlus | E-learning | - |
dc.subject.keywordPlus | End effectors | - |
dc.subject.keywordPlus | Intelligent robots | - |
dc.subject.keywordPlus | Learning systems | - |
dc.subject.keywordPlus | Neural networks | - |
dc.subject.keywordPlus | Robotics | - |
dc.subject.keywordPlus | Contextual bandits | - |
dc.subject.keywordPlus | Generalization performance | - |
dc.subject.keywordPlus | Grasp qualities | - |
dc.subject.keywordPlus | Learning-based algorithms | - |
dc.subject.keywordPlus | Online learning | - |
dc.subject.keywordPlus | Real-world problem | - |
dc.subject.keywordPlus | Robotic grasping | - |
dc.subject.keywordPlus | Shannon entropy | - |
dc.subject.keywordPlus | Learning algorithms | - |
dc.description.journalRegisteredClass | scopus | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
84, Heukseok-ro, Dongjak-gu, Seoul, Republic of Korea (06974)02-820-6194
COPYRIGHT 2019 Chung-Ang University All Rights Reserved.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.