본문

서브메뉴

Architecture, Modeling, and Optimization of Photonic Neural Network Accelerators
Architecture, Modeling, and Optimization of Photonic Neural Network Accelerators
Architecture, Modeling, and Optimization of Photonic Neural Network Accelerators

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211152834
ISBN  
9798384091998
DDC  
621.3
저자명  
Li, Shurui.
서명/저자  
Architecture, Modeling, and Optimization of Photonic Neural Network Accelerators
발행사항  
[Sl] : University of California, Los Angeles, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
255 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
주기사항  
Advisor: Gupta, Puneet.
학위논문주기  
Thesis (Ph.D.)--University of California, Los Angeles, 2024.
초록/해제  
요약As artificial intelligence (AI) continues to advance, so too must the computational architectures that support it. Traditional Complementary Metal-Oxide-Semiconductor (CMOS)- based accelerators, while having served as the backbone of computing for decades, are now encountering significant limitations due to the slowing of Moore's Law. Photonic neural network accelerators emerge as a promising alternative, offering high-speed, parallel optical computations that can significantly enhance performance and energy efficiency.Despite their potential, photonic computing systems face significant challenges, particularly related to the conversion overhead from digital to analog signals and vice versa. The conversion overhead can drastically reduce the energy efficiency of photonic neural network accelerators. This dissertation addresses these challenges by introducing cross-layer optimizations and co-design strategies that span computing methods, circuits, architectures, and algorithms. Specifically, we focus on reducing the conversion overhead through innovative design and optimization techniques.Part 1: We introduce an innovative use of free-space optical systems, specifically 4F systems, to accelerate CNNs. By leveraging Fourier optics, we reduce the complexity of convolution operations from O(N2 ) to O(N), a feat unachievable by traditional electronic systems. This part includes the design, construction, and optimization of free-space optical CNN accelerators, demonstrating significant performance improvements and energy efficiency through experimental evaluations on datasets such as MNIST and CIFAR-10.Part 2: We delve into on-chip photonic neural network accelerators, presenting two pioneering architectures: PhotoFourier and ReFOCUS. PhotoFourier leverages the Joint Transform Correlator (JTC) approach to perform convolutions with reduced complexity and fewer photonic components. ReFOCUS builds upon this with innovative features like optical buffers and wavelength-division multiplexing (WDM), further improving energy and area efficiency. We demonstrate that these on-chip designs outperform contemporary photonic and CMOS accelerators in terms of throughput, power efficiency, and energy-delay product (EDP).Part 3: We focus on algorithmic innovations and theoretical analysis to enhance the efficiency of photonic and analog computing systems. We propose a weight pool compression algorithm that reduces storage requirements and memory traffic, enabling efficient deployment of large neural networks. This compression algorithm also has a promising synergy with analog and photonic neural network accelerators, which could avoid or drastically reduce the conversion overhead of weights. Additionally, since the ADC power is heavily dependent on the bitwidth (ADC resolution), we develop a comprehensive analytical model for partial sum precision requirements, optimizing the trade-offs between accuracy and energy efficiency in analog neural network accelerators.Experimental JTC challenges and findings: We also included a chapter dedicated to the challenges and non-idealities we observed in our JTC hardware prototype. We further provide sensitivity analysis for various non-linearity of the photodetectors. The goal of this section is to complement the architecture work with some experimental analysis and findings, and provide insights and directions for future work.Through our contributions, we advance the field of photonic neural network accelerators, providing new architectures, modeling techniques, and optimization strategies. Our findings demonstrate the potential of photonic technologies to achieve high-performance, energy-efficient AI computations, thereby addressing critical challenges in modern computing hardware and paving the way for future advancements in this rapidly evolving domain.
일반주제명  
Computer engineering
일반주제명  
Condensed matter physics
일반주제명  
Computer science
일반주제명  
Information technology
키워드  
Computer architecture
키워드  
Neural network compression
키워드  
Optical neural network
키워드  
Photonic neural network accelerators
키워드  
Energy-delay product
기타저자  
University of California, Los Angeles Electrical and Computer Engineering 0333
기본자료저록  
Dissertations Abstracts International. 86-03B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017164119
■00520250211152834
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798384091998
■035    ▼a(MiAaPQ)AAI31561077
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a621.3
■1001  ▼aLi,  Shurui.
■24510▼aArchitecture,  Modeling,  and  Optimization  of  Photonic  Neural  Network  Accelerators
■260    ▼a[Sl]▼bUniversity  of  California,  Los  Angeles▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a255  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-03,  Section:  B.
■500    ▼aAdvisor:  Gupta,  Puneet.
■5021  ▼aThesis  (Ph.D.)--University  of  California,  Los  Angeles,  2024.
■520    ▼aAs  artificial  intelligence  (AI)  continues  to  advance,  so  too  must  the  computational  architectures  that  support  it.  Traditional  Complementary  Metal-Oxide-Semiconductor  (CMOS)-  based  accelerators,  while  having  served  as  the  backbone  of  computing  for  decades,  are  now  encountering  significant  limitations  due  to  the  slowing  of  Moore's  Law.  Photonic  neural  network  accelerators  emerge  as  a  promising  alternative,  offering  high-speed,  parallel  optical  computations  that  can  significantly  enhance  performance  and  energy  efficiency.Despite  their  potential,  photonic  computing  systems  face  significant  challenges,  particularly  related  to  the  conversion  overhead  from  digital  to  analog  signals  and  vice  versa.  The  conversion  overhead  can  drastically  reduce  the  energy  efficiency  of  photonic  neural  network  accelerators.  This  dissertation  addresses  these  challenges  by  introducing  cross-layer  optimizations  and  co-design  strategies  that  span  computing  methods,  circuits,  architectures,  and  algorithms.  Specifically,  we  focus  on  reducing  the  conversion  overhead  through  innovative  design  and  optimization  techniques.Part  1:  We  introduce  an  innovative  use  of  free-space  optical  systems,  specifically  4F  systems,  to  accelerate  CNNs.  By  leveraging  Fourier  optics,  we  reduce  the  complexity  of convolution  operations  from  O(N2  )  to  O(N),  a  feat  unachievable  by  traditional  electronic  systems.  This  part  includes  the  design,  construction,  and  optimization  of  free-space  optical  CNN  accelerators,  demonstrating  significant  performance  improvements  and  energy  efficiency  through  experimental  evaluations  on  datasets  such  as  MNIST  and  CIFAR-10.Part  2:  We  delve  into  on-chip  photonic  neural  network  accelerators,  presenting  two  pioneering  architectures:  PhotoFourier  and  ReFOCUS.  PhotoFourier  leverages  the  Joint  Transform  Correlator  (JTC)  approach  to  perform  convolutions  with  reduced  complexity  and  fewer  photonic  components.  ReFOCUS  builds  upon  this  with  innovative  features  like  optical  buffers  and  wavelength-division  multiplexing  (WDM),  further  improving  energy  and  area  efficiency.  We  demonstrate  that  these  on-chip  designs  outperform  contemporary  photonic  and  CMOS  accelerators  in  terms  of  throughput,  power  efficiency,  and  energy-delay  product  (EDP).Part  3:  We  focus  on  algorithmic  innovations  and  theoretical  analysis  to  enhance  the  efficiency  of  photonic  and  analog  computing  systems.  We  propose  a  weight  pool  compression  algorithm  that  reduces  storage  requirements  and  memory  traffic,  enabling  efficient  deployment  of  large  neural  networks.  This  compression  algorithm  also  has  a  promising  synergy  with  analog  and  photonic  neural  network  accelerators,  which  could  avoid  or  drastically  reduce  the  conversion  overhead  of  weights.  Additionally,  since  the  ADC  power  is  heavily  dependent  on  the  bitwidth  (ADC  resolution),  we  develop  a  comprehensive  analytical  model  for  partial  sum  precision  requirements,  optimizing  the  trade-offs  between  accuracy  and  energy  efficiency  in  analog  neural  network  accelerators.Experimental  JTC  challenges  and  findings:  We  also  included  a  chapter  dedicated  to  the  challenges  and  non-idealities  we  observed  in  our  JTC  hardware  prototype.  We  further  provide  sensitivity  analysis  for  various  non-linearity  of  the  photodetectors.  The  goal  of  this  section  is  to  complement  the  architecture  work  with  some  experimental  analysis  and  findings,  and  provide  insights  and  directions  for  future  work.Through  our  contributions,  we  advance  the  field  of  photonic  neural  network  accelerators,  providing  new  architectures,  modeling  techniques,  and  optimization  strategies.  Our  findings  demonstrate  the  potential  of  photonic  technologies  to  achieve  high-performance,  energy-efficient  AI  computations,  thereby  addressing  critical  challenges  in  modern  computing  hardware  and  paving  the  way  for  future  advancements  in  this  rapidly  evolving  domain.
■590    ▼aSchool  code:  0031.
■650  4▼aComputer  engineering
■650  4▼aCondensed  matter  physics
■650  4▼aComputer  science
■650  4▼aInformation  technology
■653    ▼aComputer  architecture
■653    ▼aNeural  network  compression
■653    ▼aOptical  neural  network
■653    ▼aPhotonic  neural  network  accelerators
■653    ▼aEnergy-delay  product
■690    ▼a0464
■690    ▼a0489
■690    ▼a0984
■690    ▼a0800
■690    ▼a0611
■71020▼aUniversity  of  California,  Los  Angeles▼bElectrical  and  Computer  Engineering  0333.
■7730  ▼tDissertations  Abstracts  International▼g86-03B.
■790    ▼a0031
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17164119▼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
    TF12012 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.