본문

서브메뉴

Scalable and Risk-Aware Verification of Learning Enabled Autonomous Systems
Scalable and Risk-Aware Verification of Learning Enabled Autonomous Systems
Scalable and Risk-Aware Verification of Learning Enabled Autonomous Systems

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211151350
ISBN  
9798382830575
DDC  
621.3
저자명  
Cleaveland, Matthew.
서명/저자  
Scalable and Risk-Aware Verification of Learning Enabled Autonomous Systems
발행사항  
[Sl] : University of Pennsylvania, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
235 p
주기사항  
Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
주기사항  
Advisor: Lee, Insup;Pappas, George J.
학위논문주기  
Thesis (Ph.D.)--University of Pennsylvania, 2024.
초록/해제  
요약As autonomous systems become more prevalent, ensuring their safety will become more and more important. However, deriving guarantees for these systems is becoming increasingly difficult due to the use of black box, learning enabled components and the growing range of operating domains in which they are deployed. The complexity of the learning-enabled components greatly increases the computational complexity of the verification problem. Additionally, the safety predictions from verifying these systems must be conservative. This thesis explores two high-level methods for verifying autonomous systems: probabilistic model checking and statistical model checking. Probabilistic model checking methods exhaustively analyze a model of the system to reason about its properties. These methods generally suffer from scalability issues, but if the abstraction is built correctly then the results will be provably conservative. On the other hand, statistical model checking methods draw traces from the system to reason about its properties. These methods don't suffer the scalability drawback of probabilistic model checking, but their guarantees are weaker and may not even be conservative. This thesis introduces methods for improving the scalability of verifying autonomous systems with probabilistic model checking methods and incorporating notions of conservatism into statistical model checking. On the probabilistic model checking side, this thesis first explores using engineering intuitions about systems to reduce probabilistic model checking complexity while preserving conservatism. Next, standard conservative probabilistic model checking techniques are used to synthesize runtime monitors that are conservative and lightweight. Finally, this thesis presents a run-time method for composing monitors of verification assumptions. Verification assumptions are critical for simplifying verification problems so that they become computationally feasible.For statistical model checking, this thesis first leverages a method called conformal prediction to bound the errors of trajectory predictors, which enables safe (i.e. conservative) planning in dynamic environments. Additionally, a method for producing less conservative conformal prediction regions in time series settings is developed. Then a method called risk verification is developed, which uses statistical methods to bound risk metrics of a system's performance. Risk metrics, which capture tail events of the system's performance, offer a statistical equivalent of worst case analysis.
일반주제명  
Electrical engineering
일반주제명  
Computer science
일반주제명  
Computer engineering
일반주제명  
Statistics
일반주제명  
Systems science
키워드  
Autonomous systems
키워드  
Computational complexity
키워드  
Scalability issues
키워드  
Probabilistic model
키워드  
Dynamic environments
기타저자  
University of Pennsylvania Electrical and Systems Engineering
기본자료저록  
Dissertations Abstracts International. 85-12B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017161395
■00520250211151350
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798382830575
■035    ▼a(MiAaPQ)AAI31243131
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a621.3
■1001  ▼aCleaveland,  Matthew.
■24510▼aScalable  and  Risk-Aware  Verification  of  Learning  Enabled  Autonomous  Systems
■260    ▼a[Sl]▼bUniversity  of  Pennsylvania▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a235  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  85-12,  Section:  B.
■500    ▼aAdvisor:  Lee,  Insup;Pappas,  George  J.
■5021  ▼aThesis  (Ph.D.)--University  of  Pennsylvania,  2024.
■520    ▼aAs  autonomous  systems  become  more  prevalent,  ensuring  their  safety  will  become  more  and  more  important.  However,  deriving  guarantees  for  these  systems  is  becoming  increasingly  difficult  due  to  the  use  of  black  box,  learning  enabled  components  and  the  growing  range  of  operating  domains  in  which  they  are  deployed.  The  complexity  of  the  learning-enabled  components  greatly  increases  the  computational  complexity  of  the  verification  problem.  Additionally,  the  safety  predictions  from  verifying  these  systems  must  be  conservative.  This  thesis  explores  two  high-level  methods  for  verifying  autonomous  systems:  probabilistic  model  checking  and  statistical  model  checking.  Probabilistic  model  checking  methods  exhaustively  analyze  a  model  of  the  system  to  reason  about  its  properties.  These  methods  generally  suffer  from  scalability  issues,  but  if  the  abstraction  is  built  correctly  then  the  results  will  be  provably  conservative.  On  the  other  hand,  statistical  model  checking  methods  draw  traces  from  the  system  to  reason  about  its  properties.  These  methods  don't  suffer  the  scalability  drawback  of  probabilistic  model  checking,  but  their  guarantees  are  weaker  and  may  not  even  be  conservative.  This  thesis  introduces  methods  for  improving  the  scalability  of  verifying  autonomous  systems  with  probabilistic  model  checking  methods  and  incorporating  notions  of  conservatism  into  statistical  model  checking. On  the  probabilistic  model  checking  side,  this  thesis  first  explores  using  engineering  intuitions  about  systems  to  reduce  probabilistic  model  checking  complexity  while  preserving  conservatism.  Next,  standard  conservative  probabilistic  model  checking  techniques  are  used  to  synthesize  runtime  monitors  that  are  conservative  and  lightweight.  Finally,  this  thesis  presents  a  run-time  method  for  composing  monitors  of  verification  assumptions.  Verification  assumptions  are  critical  for  simplifying  verification  problems  so  that  they  become  computationally  feasible.For  statistical  model  checking,  this  thesis  first  leverages  a  method  called  conformal  prediction  to  bound  the  errors  of  trajectory  predictors,  which  enables  safe  (i.e.  conservative)  planning  in  dynamic  environments.  Additionally,  a  method  for  producing  less  conservative  conformal  prediction  regions  in  time  series  settings  is  developed.  Then  a  method  called  risk  verification  is  developed,  which  uses  statistical  methods  to  bound  risk  metrics  of  a  system's  performance.  Risk  metrics,  which  capture  tail  events  of  the  system's  performance,  offer  a  statistical  equivalent  of  worst  case  analysis.
■590    ▼aSchool  code:  0175.
■650  4▼aElectrical  engineering
■650  4▼aComputer  science
■650  4▼aComputer  engineering
■650  4▼aStatistics
■650  4▼aSystems  science
■653    ▼aAutonomous  systems  
■653    ▼aComputational  complexity
■653    ▼aScalability  issues
■653    ▼aProbabilistic  model  
■653    ▼aDynamic  environments
■690    ▼a0544
■690    ▼a0984
■690    ▼a0464
■690    ▼a0463
■690    ▼a0790
■71020▼aUniversity  of  Pennsylvania▼bElectrical  and  Systems  Engineering.
■7730  ▼tDissertations  Abstracts  International▼g85-12B.
■790    ▼a0175
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17161395▼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
    TF09821 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.