본문

서브메뉴

Distribution-Free Uncertainty Quantification for Deep Learning
Distribution-Free Uncertainty Quantification for Deep Learning
Distribution-Free Uncertainty Quantification for Deep Learning

Detailed Information

자료유형  
 학위논문 서양
최종처리일시  
20260202105705
ISBN  
9798263308346
DDC  
310
저자명  
Lin, Zhen.
서명/저자  
Distribution-Free Uncertainty Quantification for Deep Learning
발행사항  
[Sl] : University of Illinois at Urbana-Champaign, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
248 p
주기사항  
Source: Dissertations Abstracts International, Volume: 87-05, Section: A.
주기사항  
Advisor: Sun, Jimeng.
학위논문주기  
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2024.
초록/해제  
요약The integration of sophisticated deep learning models into critical domains, such as healthcare, autonomous vehicles, and the legal system, is increasingly becoming a trend. These models offer significant potential for enhancing outcomes efficiently, but their adoption raises crucial challenges related to model confidence, uncertainty communication, and risk management. Uncertainty Quantification (UQ) is a key framework that addresses these issues by providing a systematic way to assess and act on the reliability of model predictions.This thesis focuses on distribution-free UQ methods, which require minimal assumptions about the data distribution and model specifics, making them highly applicable across various deep learning applications. We first investigate the problem of full calibration for deep learning classifiers, aiming to address the over-confidence or under-confidence typically observed in such classifiers. Then, we discuss methods to convert a model's output into actionable uncertainty information, expressed as prediction intervals and prediction sets. In particular, we focus on the construction of prediction intervals and sets with rigorous coverage or risk control guarantees, typically provided by conformal prediction tools. Finally, we explore the problem of UQ for natural language generation (NLG), for which we apply a graph-based approach using the similarity graph of multiple sampled responses. Through exploring model calibration, risk-controlling prediction sets, conformal prediction intervals, and UQ for NLG, this thesis aims to advance the understanding and implementation of UQ in high-stakes decision-making environments.
일반주제명  
Statistics
일반주제명  
Computer science
일반주제명  
Linguistics
키워드  
Uncertainty Quantification
키워드  
Deep learning
키워드  
Machine learning
키워드  
Conformal prediction
키워드  
Prediction sets
키워드  
Neural networks
키워드  
Healthcare applications
기타저자  
University of Illinois at Urbana-Champaign Computer Science
기본자료저록  
Dissertations Abstracts International. 87-05A.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2024        us                              c    eng  d
■001000017361096
■00520260202105705
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798263308346
■035    ▼a(MiAaPQ)AAI32409923
■035    ▼a(MiAaPQ)124231
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a310
■1001  ▼aLin,  Zhen.
■24510▼aDistribution-Free  Uncertainty  Quantification  for  Deep  Learning
■260    ▼a[Sl]▼bUniversity  of  Illinois  at  Urbana-Champaign▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a248  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  87-05,  Section:  A.
■500    ▼aAdvisor:  Sun,  Jimeng.
■5021  ▼aThesis  (Ph.D.)--University  of  Illinois  at  Urbana-Champaign,  2024.
■520    ▼aThe  integration  of  sophisticated  deep  learning  models  into  critical  domains,  such  as  healthcare,  autonomous  vehicles,  and  the  legal  system,  is  increasingly  becoming  a  trend.  These  models  offer  significant  potential  for  enhancing  outcomes  efficiently,  but  their  adoption  raises  crucial  challenges  related  to  model  confidence,  uncertainty  communication,  and  risk  management.  Uncertainty  Quantification  (UQ)  is  a  key  framework  that  addresses  these  issues  by  providing  a  systematic  way  to  assess  and  act  on  the  reliability  of  model  predictions.This  thesis  focuses  on  distribution-free  UQ  methods,  which  require  minimal  assumptions  about  the  data  distribution  and  model  specifics,  making  them  highly  applicable  across  various  deep  learning  applications.  We  first  investigate  the  problem  of  full  calibration  for  deep  learning  classifiers,  aiming  to  address  the  over-confidence  or  under-confidence  typically  observed  in  such  classifiers.  Then,  we  discuss  methods  to  convert  a  model's  output  into  actionable  uncertainty  information,  expressed  as  prediction  intervals  and  prediction  sets.  In  particular,  we  focus  on  the  construction  of  prediction  intervals  and  sets  with  rigorous  coverage  or  risk  control  guarantees,  typically  provided  by  conformal  prediction  tools.  Finally,  we  explore  the  problem  of  UQ  for  natural  language  generation  (NLG),  for  which  we  apply  a  graph-based  approach  using  the  similarity  graph  of  multiple  sampled  responses.  Through  exploring  model  calibration,  risk-controlling  prediction  sets,  conformal  prediction  intervals,  and  UQ  for  NLG,  this  thesis  aims  to  advance  the  understanding  and  implementation  of  UQ  in  high-stakes  decision-making  environments.
■590    ▼aSchool  code:  0090.
■650  4▼aStatistics
■650  4▼aComputer  science
■650  4▼aLinguistics
■653    ▼aUncertainty  Quantification
■653    ▼aDeep  learning
■653    ▼aMachine  learning
■653    ▼aConformal  prediction
■653    ▼aPrediction  sets
■653    ▼aNeural  networks
■653    ▼aHealthcare  applications
■690    ▼a0984
■690    ▼a0463
■690    ▼a0800
■690    ▼a0290
■71020▼aUniversity  of  Illinois  at  Urbana-Champaign▼bComputer  Science.
■7730  ▼tDissertations  Abstracts  International▼g87-05A.
■790    ▼a0090
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17361096▼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

    detalle info

    • Reserva
    • No existe
    • Mi carpeta
    • Primera solicitud
    • Non-Book Loan Application
    • Nighttime Book Loan Application
    Material
    número de libro número de llamada Ubicación estado Prestar info
    TF16928 전자도서 대출가능 My Folder 부재도서신고 비도서대출신청 야간 도서대출신청

    * Las reservas están disponibles en el libro de préstamos. Para hacer reservaciones, haga clic en el botón de reserva

    Books borrowed together with this book

    Related Popular Books

    Available after logging in.