본문

서브메뉴

Advances in Non-stationary Sequential Decision-Making
Advances in Non-stationary Sequential Decision-Making
Advances in Non-stationary Sequential Decision-Making

Detailed Information

자료유형  
 학위논문 서양
최종처리일시  
20250211152130
ISBN  
9798383530573
DDC  
310
저자명  
Suk, Joseph.
서명/저자  
Advances in Non-stationary Sequential Decision-Making
발행사항  
[Sl] : Columbia University, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
359 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-01, Section: B.
주기사항  
Advisor: Kpotufe, Samory.
학위논문주기  
Thesis (Ph.D.)--Columbia University, 2024.
초록/해제  
요약We study the problem of sequential decision-making (e.g. multi-armed bandits, contextual bandits, reinforcement learning) under changing environments, or distribution shifts. Ideally, one aims to automatically adapt/self-tune to unknown changes in distribution, and restart exploration as needed. While recent theoretical breakthroughs show this is possible in a broad sense, such works contend that the learner should restart procedures upon experiencing any change leading to worst-case (regret) rates. This leaves open whether faster rates are possible, adaptively, if few changes in distribution are actually severe, e.g., involve no change in best action.This thesis initiates a broad research program giving positive answers to these open questions across several instances. In particular, we begin at non-stationary bandits and show a much weaker notion of change can be adapted to, which can yield significantly faster rates than previously known, whether as expressed in terms of number of best action switches-for which no adaptive procedure was known, or in terms of previously studied variation or smoothness measures. We then generalize these results to non-parametric contextual bandits and dueling bandits. As a result, we substantially improve the theoretical state-of-the-art performance guarantees for these problems and, in many cases, tightly characterize the statistical limits of sequential decision-making under changing environments.
일반주제명  
Statistics
일반주제명  
Computer science
키워드  
Minimax
키워드  
Multi-armed bandits
키워드  
Non-parametric statistics
키워드  
Statistical learning theory
키워드  
Reinforcement learning
기타저자  
Columbia University Statistics
기본자료저록  
Dissertations Abstracts International. 86-01B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017163056
■00520250211152130
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798383530573
■035    ▼a(MiAaPQ)AAI31483147
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a310
■1001  ▼aSuk,  Joseph.
■24510▼aAdvances  in  Non-stationary  Sequential  Decision-Making
■260    ▼a[Sl]▼bColumbia  University▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a359  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-01,  Section:  B.
■500    ▼aAdvisor:  Kpotufe,  Samory.
■5021  ▼aThesis  (Ph.D.)--Columbia  University,  2024.
■520    ▼aWe  study  the  problem  of  sequential  decision-making  (e.g.  multi-armed  bandits,  contextual  bandits,  reinforcement  learning)  under  changing  environments,  or  distribution  shifts.  Ideally,  one  aims  to  automatically  adapt/self-tune  to  unknown  changes  in  distribution,  and  restart  exploration  as  needed.  While  recent  theoretical  breakthroughs  show  this  is  possible  in  a  broad  sense,  such  works  contend  that  the  learner  should  restart  procedures  upon  experiencing  any  change  leading  to  worst-case  (regret)  rates.  This  leaves  open  whether  faster  rates  are  possible,  adaptively,  if  few  changes  in  distribution  are  actually  severe,  e.g.,  involve  no  change  in  best  action.This  thesis  initiates  a  broad  research  program  giving  positive  answers  to  these  open  questions  across  several  instances.  In  particular,  we  begin  at  non-stationary  bandits  and  show  a  much  weaker  notion  of  change  can  be  adapted  to,  which  can  yield  significantly  faster  rates  than  previously  known,  whether  as  expressed  in  terms  of  number  of  best  action  switches-for  which  no  adaptive  procedure  was  known,  or  in  terms  of  previously  studied  variation  or  smoothness  measures.  We  then  generalize  these  results  to  non-parametric  contextual  bandits  and  dueling  bandits.  As  a  result,  we  substantially  improve  the  theoretical  state-of-the-art  performance  guarantees  for  these  problems  and,  in  many  cases,  tightly  characterize  the  statistical  limits  of  sequential  decision-making  under  changing  environments.
■590    ▼aSchool  code:  0054.
■650  4▼aStatistics
■650  4▼aComputer  science
■653    ▼aMinimax
■653    ▼aMulti-armed  bandits
■653    ▼aNon-parametric  statistics
■653    ▼aStatistical  learning  theory
■653    ▼aReinforcement  learning
■690    ▼a0463
■690    ▼a0984
■690    ▼a0800
■71020▼aColumbia  University▼bStatistics.
■7730  ▼tDissertations  Abstracts  International▼g86-01B.
■790    ▼a0054
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17163056▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

Preview

Export

ChatGPT Discussion

AI Recommended Related Books


    New Books MORE
    Statistics for the past 3 years. Go to brief

    Подробнее информация.

    • Бронирование
    • не существует
    • моя папка
    • Первый запрос зрения
    • Non-Book Loan Application
    • Nighttime Book Loan Application
    материал
    Reg No. Количество платежных Местоположение статус Ленд информации
    TF10946 전자도서 대출가능 My Folder 부재도서신고 비도서대출신청 야간 도서대출신청

    * Бронирование доступны в заимствований книги. Чтобы сделать предварительный заказ, пожалуйста, нажмите кнопку бронирование

    Books borrowed together with this book

    Related Popular Books

    Available after logging in.