본문

서브메뉴

New Perspectives on Adversarially Robust Machine Learning Systems
New Perspectives on Adversarially Robust Machine Learning Systems
New Perspectives on Adversarially Robust Machine Learning Systems

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211151009
ISBN  
9798384447146
DDC  
004
저자명  
Sitawarin, Chawin.
서명/저자  
New Perspectives on Adversarially Robust Machine Learning Systems
발행사항  
[Sl] : University of California, Berkeley, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
170 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
주기사항  
Advisor: Wagner, David.
학위논문주기  
Thesis (Ph.D.)--University of California, Berkeley, 2024.
초록/해제  
요약Security has always been at the core of computer systems, from hardware to software, and network. Through their recent advancement, machine learning and artificial intelligence have found themselves an essential space in this software stack. This shiny new addition pushes the boundary of computer programs beyond what humans have imagined, from multi-media editing to intelligent personal assistants. Unfortunately, it is also becoming the weakest security link in this stack. One of the most alarming concerns of these ML and in particular, deep learning systems is the lack of robustness, a phenomenon termed adversarial examples.This Ph.D. dissertation presents an in-depth investigation into the adversarial robustness of deep learning systems with the goal of building a practical defense against these attacks. It consists of three main parts. The first focuses on improving the state-of-the-art defense, adversarial training, by means of high-quality data and supervision. We show that fine-grained supervision during training can increase the robustness of neural networks on an object classification task. In the second part, we take on a broader and more practical perspective on the defenses. We argue that the model-level defense, i.e., building more adversarially robust models, alone is necessary but not sufficient to achieve a secure system in practice. Instead, we propose a new model-level defense that when combined with the existing system-level defense, can provide a practical solution to an important and realistic type of attack. While our method does not completely stop all adversarial attacks, it shows that building a "reasonably" secure ML system may be within closer reach than the community largely believes. In the final part of this dissertation, we demonstrate a novel practical attack algorithm against a real-world large language model API with little cost and no human intervention. Identifying vulnerabilities is the first step to solving them. We hope that the insights developed in this dissertation will provide new perspectives to the research community and play an instrumental role in building a secure system against adversarial examples.
일반주제명  
Computer science
일반주제명  
Information technology
키워드  
Adversarial examples
키워드  
Adversarial robustness
키워드  
Deep learning
키워드  
Machine learning
키워드  
System-level defense
기타저자  
University of California, Berkeley Computer Science
기본자료저록  
Dissertations Abstracts International. 86-03B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017160387
■00520250211151009
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798384447146
■035    ▼a(MiAaPQ)AAI30995220
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a004
■1001  ▼aSitawarin,  Chawin.
■24510▼aNew  Perspectives  on  Adversarially  Robust  Machine  Learning  Systems
■260    ▼a[Sl]▼bUniversity  of  California,  Berkeley▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a170  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-03,  Section:  B.
■500    ▼aAdvisor:  Wagner,  David.
■5021  ▼aThesis  (Ph.D.)--University  of  California,  Berkeley,  2024.
■520    ▼aSecurity  has  always  been  at  the  core  of  computer  systems,  from  hardware  to  software,  and  network.  Through  their  recent  advancement,  machine  learning  and  artificial  intelligence  have  found  themselves  an  essential  space  in  this  software  stack.  This  shiny  new  addition  pushes  the  boundary  of  computer  programs  beyond  what  humans  have  imagined,  from  multi-media  editing  to  intelligent  personal  assistants.  Unfortunately,  it  is  also  becoming  the  weakest  security  link  in  this  stack.  One  of  the  most  alarming  concerns  of  these  ML  and  in  particular,  deep  learning  systems  is  the  lack  of  robustness,  a  phenomenon  termed  adversarial  examples.This  Ph.D.  dissertation  presents  an  in-depth  investigation  into  the  adversarial  robustness  of  deep  learning  systems  with  the  goal  of  building  a  practical  defense  against  these  attacks.  It  consists  of  three  main  parts.  The  first  focuses  on  improving  the  state-of-the-art  defense,  adversarial  training,  by  means  of  high-quality  data  and  supervision.  We  show  that  fine-grained  supervision  during  training  can  increase  the  robustness  of  neural  networks  on  an  object  classification  task.  In  the  second  part,  we  take  on  a  broader  and  more  practical  perspective  on  the  defenses.  We  argue  that  the  model-level  defense,  i.e.,  building  more  adversarially  robust  models,  alone  is  necessary  but  not  sufficient  to  achieve  a  secure  system  in  practice.  Instead,  we  propose  a  new  model-level  defense  that  when  combined  with  the  existing  system-level  defense,  can  provide  a  practical  solution  to  an  important  and  realistic  type  of  attack.  While  our  method  does  not  completely  stop  all  adversarial  attacks,  it  shows  that  building  a  "reasonably"  secure  ML  system  may  be  within  closer  reach  than  the  community  largely  believes.  In  the  final  part  of  this  dissertation,  we  demonstrate  a  novel  practical  attack  algorithm  against  a  real-world  large  language  model  API  with  little  cost  and  no  human  intervention.  Identifying  vulnerabilities  is  the  first  step  to  solving  them.  We  hope  that  the  insights  developed  in  this  dissertation  will  provide  new  perspectives  to  the  research  community  and  play  an  instrumental  role  in  building  a  secure  system  against  adversarial  examples.
■590    ▼aSchool  code:  0028.
■650  4▼aComputer  science
■650  4▼aInformation  technology
■653    ▼aAdversarial  examples
■653    ▼aAdversarial  robustness
■653    ▼aDeep  learning
■653    ▼aMachine  learning
■653    ▼aSystem-level  defense
■690    ▼a0984
■690    ▼a0489
■690    ▼a0800
■71020▼aUniversity  of  California,  Berkeley▼bComputer  Science.
■7730  ▼tDissertations  Abstracts  International▼g86-03B.
■790    ▼a0028
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17160387▼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
    TF13141 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.