본문

서브메뉴

Algorithm-Hardware Co-Design for Deep Learning and Probabilistic Computing with Compute-In-Memory Accelerators
Algorithm-Hardware Co-Design for Deep Learning and Probabilistic Computing with Compute-In...
Algorithm-Hardware Co-Design for Deep Learning and Probabilistic Computing with Compute-In-Memory Accelerators

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20260202105601
ISBN  
9798263394080
DDC  
620
저자명  
Lu, Anni.
서명/저자  
Algorithm-Hardware Co-Design for Deep Learning and Probabilistic Computing with Compute-In-Memory Accelerators
발행사항  
[Sl] : Georgia Institute of Technology, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
117 p
주기사항  
Source: Dissertations Abstracts International, Volume: 87-05, Section: A.
주기사항  
Advisor: Yu, Shimeng.
학위논문주기  
Thesis (Ph.D.)--Georgia Institute of Technology, 2024.
초록/해제  
요약Compute-in-memory (CIM) is an attractive solution to accelerate the multiplyaccumulate (MAC) operations of deep learning and beyond due to its high processing parallelism and energy efficiency. In this thesis, algorithm-hardware co-optimizations of CIM are explored across diverse models and applications. First, the integrated benchmark framework "NeuroSim" for CIM accelerators is validated with actual silicon data and calibrated with adjustment factors. Based on this simulator, the CIM systems for deep neural network (DNN) with limited on-chip resources are explored, addressing the challenge of accommodating large-scale models on area-constrained CIM chips, and the reconfiguration of deploying different models on prefabricated CIM chip with fixed hardware resources.The CIM scheme is further extended to probabilistic computing, where the memory and circuit intrinsic stochasticity are no longer harmful to model accuracy but provides energy-efficient random number generation. A novel CIM accelerator for Bayesian neural network is proposed to generate Gaussian distributed weights using the probabilistic switching of spin-orbit torque magnetic random-access memory (SOT-MRAM) in weak programming. The inherent memory noise is also utilized for scalable and clustered inmemory Ising annealers to solve NP-hard combinatorial optimization problems. An analog CIM annealer using temporal variation of charge-trap transistor and a digital CIM annealer using process variation of static random-access memory (SRAM) are proposed. Design space explorations and system-level performance evaluations are also performed by modifying the validated NeuroSim simulator.
일반주제명  
Random access memory
일반주제명  
Deep learning
일반주제명  
Semiconductors
일반주제명  
Recommender systems
일반주제명  
Normal distribution
일반주제명  
Circuits
일반주제명  
CMOS
일반주제명  
Co-design
일반주제명  
Energy consumption
일반주제명  
Field programmable gate arrays
일반주제명  
Annealing
일반주제명  
Space exploration
일반주제명  
Neural networks
일반주제명  
Energy efficiency
일반주제명  
Aerospace engineering
일반주제명  
Computer science
일반주제명  
Electrical engineering
일반주제명  
Industrial engineering
일반주제명  
Information science
일반주제명  
Sustainability
기타저자  
Georgia Institute of Technology.
기본자료저록  
Dissertations Abstracts International. 87-05A.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2024        us                              c    eng  d
■001000017360649
■00520260202105601
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798263394080
■035    ▼a(MiAaPQ)AAI32315975
■035    ▼a(MiAaPQ)GeorgiaTech75293
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a620
■1001  ▼aLu,  Anni.
■24510▼aAlgorithm-Hardware  Co-Design  for  Deep  Learning  and  Probabilistic  Computing  with  Compute-In-Memory  Accelerators
■260    ▼a[Sl]▼bGeorgia  Institute  of  Technology▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a117  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  87-05,  Section:  A.
■500    ▼aAdvisor:  Yu,  Shimeng.
■5021  ▼aThesis  (Ph.D.)--Georgia  Institute  of  Technology,  2024.
■520    ▼aCompute-in-memory  (CIM)  is  an  attractive  solution  to  accelerate  the  multiplyaccumulate  (MAC)  operations  of  deep  learning  and  beyond  due  to  its  high  processing  parallelism  and  energy  efficiency.  In  this  thesis,  algorithm-hardware  co-optimizations  of  CIM  are  explored  across  diverse  models  and  applications.  First,  the  integrated  benchmark  framework  "NeuroSim"  for  CIM  accelerators  is  validated  with  actual  silicon  data  and  calibrated  with  adjustment  factors.  Based  on  this  simulator,  the  CIM  systems  for  deep  neural  network  (DNN)  with  limited  on-chip  resources  are  explored,  addressing  the  challenge  of  accommodating  large-scale  models  on  area-constrained  CIM  chips,  and  the  reconfiguration  of  deploying  different  models  on  prefabricated  CIM  chip  with  fixed  hardware  resources.The  CIM  scheme  is  further  extended  to  probabilistic  computing,  where  the  memory  and  circuit  intrinsic  stochasticity  are  no  longer  harmful  to  model  accuracy  but  provides  energy-efficient  random  number  generation.  A  novel  CIM  accelerator  for  Bayesian  neural  network  is  proposed  to  generate  Gaussian  distributed  weights  using  the  probabilistic  switching  of  spin-orbit  torque  magnetic  random-access  memory  (SOT-MRAM)  in  weak  programming.  The  inherent  memory  noise  is  also  utilized  for  scalable  and  clustered  inmemory  Ising  annealers  to  solve  NP-hard  combinatorial  optimization  problems.  An  analog  CIM  annealer  using  temporal  variation  of  charge-trap  transistor  and  a  digital  CIM  annealer  using  process  variation  of  static  random-access  memory  (SRAM)  are  proposed.  Design  space  explorations  and  system-level  performance  evaluations  are  also  performed  by  modifying  the  validated  NeuroSim  simulator.
■590    ▼aSchool  code:  0078.
■650  4▼aRandom  access  memory
■650  4▼aDeep  learning
■650  4▼aSemiconductors
■650  4▼aRecommender  systems
■650  4▼aNormal  distribution
■650  4▼aCircuits
■650  4▼aCMOS
■650  4▼aCo-design
■650  4▼aEnergy  consumption
■650  4▼aField  programmable  gate  arrays
■650  4▼aAnnealing
■650  4▼aSpace  exploration
■650  4▼aNeural  networks
■650  4▼aEnergy  efficiency
■650  4▼aAerospace  engineering
■650  4▼aComputer  science
■650  4▼aElectrical  engineering
■650  4▼aIndustrial  engineering
■650  4▼aInformation  science
■650  4▼aSustainability
■690    ▼a0538
■690    ▼a0800
■690    ▼a0984
■690    ▼a0544
■690    ▼a0546
■690    ▼a0723
■690    ▼a0640
■71020▼aGeorgia  Institute  of  Technology.
■7730  ▼tDissertations  Abstracts  International▼g87-05A.
■790    ▼a0078
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360649▼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
    TF17456 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.