서브메뉴
검색
Adaptive Sequential Decision Making: Bandit Optimization and Active Learning- [electronic resource]
Adaptive Sequential Decision Making: Bandit Optimization and Active Learning- [electronic resource]
상세정보
- 자료유형
- 학위논문
- Control Number
- 0016935006
- International Standard Book Number
- 9798380616409
- Dewey Decimal Classification Number
- 004
- Main Entry-Personal Name
- Liu, Chong.
- Publication, Distribution, etc. (Imprint
- [S.l.] : University of California, Santa Barbara., 2023
- Publication, Distribution, etc. (Imprint
- Ann Arbor : ProQuest Dissertations & Theses, 2023
- Physical Description
- 1 online resource(194 p.)
- General Note
- Source: Dissertations Abstracts International, Volume: 85-04, Section: B.
- General Note
- Advisor: Wang, Yu-Xiang.
- Dissertation Note
- Thesis (Ph.D.)--University of California, Santa Barbara, 2023.
- Restrictions on Access Note
- This item must not be sold to any third party vendors.
- Summary, Etc.
- 요약Deep neural networks usually have many hyperparameters that need to be tuned. Modern material design problems usually require material scientists to sequentially select processing parameters and conduct experiments to observe material performances. To save privacy cost, the learning system needs to carefully choose queries to answer under the differential privacy framework. To train a robot under video guidance, engineers need to carefully choose video samples for training. However, in all cases, people cannot observe performances of unselected actions and the experimental cost can be huge. These two challenges hinder efficient neural network training, new material design, privacy protection, and robot training and call for actions. In this thesis, I present my research on optimization, bandits, and active learning under the adaptive sequential decision making framework. My algorithms are able to solve black box function optimization without the curse of dimensionality, achieve no regret under the function class misspecification, reduce privacy cost under the differential privacy framework, and significantly reduce video sample complexity for robot training. All of them come with theoretical or empirical analysis.
- Subject Added Entry-Topical Term
- Computer science.
- Subject Added Entry-Topical Term
- Information technology.
- Index Term-Uncontrolled
- Deep neural networks
- Index Term-Uncontrolled
- Privacy framework
- Index Term-Uncontrolled
- Privacy protection
- Index Term-Uncontrolled
- Decision making
- Index Term-Uncontrolled
- Bandit optimization
- Added Entry-Corporate Name
- University of California, Santa Barbara Computer Science
- Host Item Entry
- Dissertations Abstracts International. 85-04B.
- Host Item Entry
- Dissertation Abstract International
- Electronic Location and Access
- 로그인을 한후 보실 수 있는 자료입니다.
- Control Number
- joongbu:640727
MARC
008240220s2023 ulk 00 kor■001000016935006
■00520240214101842
■006m o d
■007cr#unu||||||||
■020 ▼a9798380616409
■035 ▼a(MiAaPQ)AAI30638337
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a004
■1001 ▼aLiu, Chong.
■24510▼aAdaptive Sequential Decision Making: Bandit Optimization and Active Learning▼h[electronic resource]
■260 ▼a[S.l.]▼bUniversity of California, Santa Barbara. ▼c2023
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2023
■300 ▼a1 online resource(194 p.)
■500 ▼aSource: Dissertations Abstracts International, Volume: 85-04, Section: B.
■500 ▼aAdvisor: Wang, Yu-Xiang.
■5021 ▼aThesis (Ph.D.)--University of California, Santa Barbara, 2023.
■506 ▼aThis item must not be sold to any third party vendors.
■520 ▼aDeep neural networks usually have many hyperparameters that need to be tuned. Modern material design problems usually require material scientists to sequentially select processing parameters and conduct experiments to observe material performances. To save privacy cost, the learning system needs to carefully choose queries to answer under the differential privacy framework. To train a robot under video guidance, engineers need to carefully choose video samples for training. However, in all cases, people cannot observe performances of unselected actions and the experimental cost can be huge. These two challenges hinder efficient neural network training, new material design, privacy protection, and robot training and call for actions. In this thesis, I present my research on optimization, bandits, and active learning under the adaptive sequential decision making framework. My algorithms are able to solve black box function optimization without the curse of dimensionality, achieve no regret under the function class misspecification, reduce privacy cost under the differential privacy framework, and significantly reduce video sample complexity for robot training. All of them come with theoretical or empirical analysis.
■590 ▼aSchool code: 0035.
■650 4▼aComputer science.
■650 4▼aInformation technology.
■653 ▼aDeep neural networks
■653 ▼aPrivacy framework
■653 ▼aPrivacy protection
■653 ▼aDecision making
■653 ▼aBandit optimization
■690 ▼a0984
■690 ▼a0489
■690 ▼a0800
■71020▼aUniversity of California, Santa Barbara▼bComputer Science.
■7730 ▼tDissertations Abstracts International▼g85-04B.
■773 ▼tDissertation Abstract International
■790 ▼a0035
■791 ▼aPh.D.
■792 ▼a2023
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T16935006▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
■980 ▼a202402▼f2024
