본문

서브메뉴

Essays in Methodology
Essays in Methodology
Essays in Methodology

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20260202104846
ISBN  
9798293893201
DDC  
310
저자명  
Bouyamourn, Adam.
서명/저자  
Essays in Methodology
발행사항  
[Sl] : University of California, Berkeley, 2025
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2025
형태사항  
163 p
주기사항  
Source: Dissertations Abstracts International, Volume: 87-04, Section: B.
주기사항  
Advisor: Dunning, Thad.
학위논문주기  
Thesis (Ph.D.)--University of California, Berkeley, 2025.
초록/해제  
요약This dissertation studies three problems in statistical methodology.How should researchers select experimental sites when the deployment population may differ from observed data? The first paper, "Site Selection under Distribution shift via Optimal Transport and Wasserstein Distributionally-Robust Optimization'', formulates the problem of experimental site selection as an optimal transport problem, and develops methods to minimize downstream estimation error by choosing sites that minimize Wasserstein distances between population and sample covariate distributions. I develop new theoretical upper bounds on PATE and CATE estimation errors, and show that these different objectives lead to different site selection strategies: both approaches form balanced, representative partitions of the support of the covariates, but use different penalties, which place a larger emphasis on site selections that place greater weight on minimizing downstream bias (PATE) and and variance (CATE) respectively. I extend this approach by using Wasserstein Distributionally Robust Optimization to guard against distribution shift when observed sites may not represent the target population, and develop a novel, data-driven procedure for uncertainty radius selection. I develop a cutting-plane algorithm that solves the resulting minimax problem by exploiting its sequential game structure, combined with a data-adaptive procedure for calibrating robustness parameters without requiring arbitrary assumptions about distributional uncertainty. Simulation evidence, and a reanalysis of a randomized microcredit experiment in Morocco, show that these methods outperform random and stratified sampling of sites, and alternative optimization methods i) for moderate-to-large size problem instances ii) when covariates are moderately informative about treatment effects, and iii) under induced distribution shift.The second paper, "Collusive and Adversarial Replication'', studies a game in which social ties between members of a research community may discourage prospective replicators from debunking papers that misreport results. Here, replication is an entrance decision, as a Replicator chooses whether or not to Replicate a given paper. A high level of social connectedness between members of a research community increases the field-wise False Discovery Rate, a measure of the social welfare associated with a healthy publication process. The moral is that larger, more diverse academic fields with fewer social ties are more likely to have an adversarial culture around replication, and that this improves social welfare. I consider three proposals to improve replication practices: Random auditing, or police-patrol replication; automated unit tests; and a recent proposal to lower the threshold for statistical significance. I argue that random auditing and automated unit tests can improve social welfare, but that the effect of lowering the statistical significance threshold is ambiguous.The third paper, "Why LLMs Hallucinate'', argues that LLMs hallucinate because their output is not constrained to be synonymous with claims for which they have evidence: a condition I call evidential closure. Information about the truth or falsity of sentences is not statistically identified in the standard neural language generation setup, and so cannot be conditioned on to generate new strings. We then show how to constrain LLMs to produce output that satisfies evidential closure. A multimodal LLM must learn about the external world (perceptual learning); it must learn a mapping from strings to states of the world (extensional learning); and, to achieve fluency when generalizing beyond a body of evidence, it must learn mappings from strings to their synonyms (intensional learning). The output of a unimodal LLM must be synonymous with strings in a validated evidence set. Finally, I present a heuristic procedure, Learn-Babble-Prune, that yields faithful output from an LLM by rejecting output that is not synonymous with claims for which the LLM has evidence.
일반주제명  
Statistics
일반주제명  
Political science
일반주제명  
Computer science
키워드  
Causal inference
키워드  
Selective inference
키워드  
Perceptual learning
기타저자  
University of California, Berkeley Political Science
기본자료저록  
Dissertations Abstracts International. 87-04B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2025        us                              c    eng  d
■001000017359187
■00520260202104846
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798293893201
■035    ▼a(MiAaPQ)AAI32173835
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a310
■1001  ▼aBouyamourn,  Adam.
■24510▼aEssays  in  Methodology
■260    ▼a[Sl]▼bUniversity  of  California,  Berkeley▼c2025
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2025
■300    ▼a163  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  87-04,  Section:  B.
■500    ▼aAdvisor:  Dunning,  Thad.
■5021  ▼aThesis  (Ph.D.)--University  of  California,  Berkeley,  2025.
■520    ▼aThis  dissertation  studies  three  problems  in  statistical  methodology.How  should  researchers  select  experimental  sites  when  the  deployment  population  may  differ  from  observed  data?  The  first  paper,  "Site  Selection  under  Distribution  shift  via  Optimal  Transport  and  Wasserstein  Distributionally-Robust  Optimization'',  formulates  the  problem  of  experimental  site  selection  as  an  optimal  transport  problem,  and  develops  methods  to  minimize  downstream  estimation  error  by  choosing  sites  that  minimize  Wasserstein  distances  between  population  and  sample  covariate  distributions.  I  develop  new  theoretical  upper  bounds  on  PATE  and  CATE  estimation  errors,  and  show  that  these  different  objectives  lead  to  different  site  selection  strategies:  both  approaches  form  balanced,  representative  partitions  of  the  support  of  the  covariates,  but  use  different  penalties,  which  place  a  larger  emphasis  on  site  selections  that  place  greater  weight  on  minimizing  downstream  bias  (PATE)  and  and  variance  (CATE)  respectively.  I  extend  this  approach  by  using  Wasserstein  Distributionally  Robust  Optimization  to  guard  against  distribution  shift  when  observed  sites  may  not  represent  the  target  population,  and  develop  a  novel,  data-driven  procedure  for  uncertainty  radius  selection.  I  develop  a  cutting-plane  algorithm  that  solves  the  resulting  minimax  problem  by  exploiting  its  sequential  game  structure,  combined  with  a  data-adaptive  procedure  for  calibrating  robustness  parameters  without  requiring  arbitrary  assumptions  about  distributional  uncertainty.  Simulation  evidence,  and  a  reanalysis  of  a  randomized  microcredit  experiment  in  Morocco,  show  that  these  methods  outperform  random  and  stratified  sampling  of  sites,  and  alternative  optimization  methods  i)  for  moderate-to-large  size  problem  instances  ii)  when  covariates  are  moderately  informative  about  treatment  effects,  and  iii)  under  induced  distribution  shift.The  second  paper,  "Collusive  and  Adversarial  Replication'',  studies  a  game  in  which  social  ties  between  members  of  a  research  community  may  discourage  prospective  replicators  from  debunking  papers  that  misreport  results.  Here,  replication  is  an  entrance  decision,  as  a  Replicator  chooses  whether  or  not  to  Replicate  a  given  paper.  A  high  level  of  social  connectedness  between  members  of  a  research  community  increases  the  field-wise  False  Discovery  Rate,  a  measure  of  the  social  welfare  associated  with  a  healthy  publication  process.  The  moral  is  that  larger,  more  diverse  academic  fields  with  fewer  social  ties  are  more  likely  to  have  an  adversarial  culture  around  replication,  and  that  this  improves  social  welfare.  I  consider  three  proposals  to  improve  replication  practices:  Random  auditing,  or  police-patrol  replication;  automated  unit  tests;  and  a  recent  proposal  to  lower  the  threshold  for  statistical  significance.  I  argue  that  random  auditing  and  automated  unit  tests  can  improve  social  welfare,  but  that  the  effect  of  lowering  the  statistical  significance  threshold  is  ambiguous.The  third  paper,  "Why  LLMs  Hallucinate'',  argues  that  LLMs  hallucinate  because  their  output  is  not  constrained  to  be  synonymous  with  claims  for  which  they  have  evidence:  a  condition  I  call  evidential  closure.  Information  about  the  truth  or  falsity  of  sentences  is  not  statistically  identified  in  the  standard  neural  language  generation  setup,  and  so  cannot  be  conditioned  on  to  generate  new  strings.  We  then  show  how  to  constrain  LLMs  to  produce  output  that  satisfies  evidential  closure.  A  multimodal  LLM  must  learn  about  the  external  world  (perceptual  learning);  it  must  learn  a  mapping  from  strings  to  states  of  the  world  (extensional  learning);  and,  to  achieve  fluency  when  generalizing  beyond  a  body  of  evidence,  it  must  learn  mappings  from  strings  to  their  synonyms  (intensional  learning).  The  output  of  a  unimodal  LLM  must  be  synonymous  with  strings  in  a  validated  evidence  set.  Finally,  I  present  a  heuristic  procedure,  Learn-Babble-Prune,  that  yields  faithful  output  from  an  LLM  by  rejecting  output  that  is  not  synonymous  with  claims  for  which  the  LLM  has  evidence.
■590    ▼aSchool  code:  0028.
■650  4▼aStatistics
■650  4▼aPolitical  science
■650  4▼aComputer  science
■653    ▼aCausal  inference
■653    ▼aSelective  inference
■653    ▼aPerceptual  learning
■690    ▼a0463
■690    ▼a0615
■690    ▼a0984
■71020▼aUniversity  of  California,  Berkeley▼bPolitical  Science.
■7730  ▼tDissertations  Abstracts  International▼g87-04B.
■790    ▼a0028
■791    ▼aPh.D.
■792    ▼a2025
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17359187▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

미리보기

내보내기

chatGPT토론

Ai 추천 관련 도서


    신착도서 더보기
    최근 3년간 통계입니다.

    소장정보

    • 예약
    • 소재불명신고
    • 나의폴더
    • 우선정리요청
    • 비도서대출신청
    • 야간 도서대출신청
    소장자료
    등록번호 청구기호 소장처 대출가능여부 대출정보
    TF14714 전자도서 대출가능 마이폴더 부재도서신고 비도서대출신청 야간 도서대출신청

    * 대출중인 자료에 한하여 예약이 가능합니다. 예약을 원하시면 예약버튼을 클릭하십시오.

    해당 도서를 다른 이용자가 함께 대출한 도서

    관련 인기도서

    로그인 후 이용 가능합니다.