서브메뉴
검색
Early-Bird Training and Hierarchical Hardware-Aware Model Compression Towards Green and Ubiquitous Artificial Intelligence
Early-Bird Training and Hierarchical Hardware-Aware Model Compression Towards Green and Ubiquitous Artificial Intelligence
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105533
- ISBN
- 9798263346386
- DDC
- 330
- 저자명
- You, Haoran.
- 서명/저자
- Early-Bird Training and Hierarchical Hardware-Aware Model Compression Towards Green and Ubiquitous Artificial Intelligence
- 발행사항
- [Sl] : Georgia Institute of Technology, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 155 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-05, Section: B.
- 주기사항
- Advisor: Lin, Yingyan.
- 학위논문주기
- Thesis (Ph.D.)--Georgia Institute of Technology, 2025.
- 초록/해제
- 요약Artificial intelligence (AI) has made remarkable breakthroughs across various applications, such as image perception, augmented and virtual reality (AR/VR), and AI-generated content (AIGC). However, critical research gaps remain between the powerful yet large-scale AI models and the computational constraints of both edge and cloud platforms, including (1) the efficiency gap, which hinders the acceleration of AI training and the ability to iterate quickly; (2) the scalability gap, which challenges efficient scaling and deployment of AI models on cloud GPUs; and (3) the accessibility gap, which limits the feasibility of running small-scale AI models on resource-constrained edge devices like smartphones, AR/VR headsets, and IoT sensors. In this thesis, I will introduce three key strategies to enable efficient, scalable, and accessible AI across cloud and edge. First, I will introduce Early-Bird Tickets, a method that identifies efficient subnetworks from a large AI model during early training stages, achieving 5~10x training efficiency. Second, I will present ShiftAddNet, a hardware-aware AI algorithm that replaces costly multiplications with hardware-efficient shift and add operators, improving scalability and efficiency for large-scale vision and language models. Third, I will advocate for a holistic AI system co-design approach that optimizes both algorithms and hardware. For example, I will showcase ViTCoD, the first Vision Transformer (ViT) algorithm-hardware co-design framework that leverages the unique characteristics of ViTs to boost system performance. By combining these approaches, this thesis enables efficient, scalable, and accessible AI training and deployment, closing the three gaps towards ubiquitous AI across cloud and edge platforms.
- 일반주제명
- Sparsity
- 일반주제명
- Co-design
- 일반주제명
- Visualization
- 일반주제명
- Large language models
- 기본자료저록
- Dissertations Abstracts International. 87-05B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017360478
■00520260202105533
■006m o d
■007cr#unu||||||||
■020 ▼a9798263346386
■035 ▼a(MiAaPQ)AAI32309931
■035 ▼a(MiAaPQ)GeorgiaTech77931
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a330
■1001 ▼aYou, Haoran.
■24510▼aEarly-Bird Training and Hierarchical Hardware-Aware Model Compression Towards Green and Ubiquitous Artificial Intelligence
■260 ▼a[Sl]▼bGeorgia Institute of Technology▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a155 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-05, Section: B.
■500 ▼aAdvisor: Lin, Yingyan.
■5021 ▼aThesis (Ph.D.)--Georgia Institute of Technology, 2025.
■520 ▼aArtificial intelligence (AI) has made remarkable breakthroughs across various applications, such as image perception, augmented and virtual reality (AR/VR), and AI-generated content (AIGC). However, critical research gaps remain between the powerful yet large-scale AI models and the computational constraints of both edge and cloud platforms, including (1) the efficiency gap, which hinders the acceleration of AI training and the ability to iterate quickly; (2) the scalability gap, which challenges efficient scaling and deployment of AI models on cloud GPUs; and (3) the accessibility gap, which limits the feasibility of running small-scale AI models on resource-constrained edge devices like smartphones, AR/VR headsets, and IoT sensors. In this thesis, I will introduce three key strategies to enable efficient, scalable, and accessible AI across cloud and edge. First, I will introduce Early-Bird Tickets, a method that identifies efficient subnetworks from a large AI model during early training stages, achieving 5~10x training efficiency. Second, I will present ShiftAddNet, a hardware-aware AI algorithm that replaces costly multiplications with hardware-efficient shift and add operators, improving scalability and efficiency for large-scale vision and language models. Third, I will advocate for a holistic AI system co-design approach that optimizes both algorithms and hardware. For example, I will showcase ViTCoD, the first Vision Transformer (ViT) algorithm-hardware co-design framework that leverages the unique characteristics of ViTs to boost system performance. By combining these approaches, this thesis enables efficient, scalable, and accessible AI training and deployment, closing the three gaps towards ubiquitous AI across cloud and edge platforms.
■590 ▼aSchool code: 0078.
■650 4▼aSparsity
■650 4▼aCo-design
■650 4▼aVisualization
■650 4▼aLarge language models
■690 ▼a0800
■71020▼aGeorgia Institute of Technology.
■7730 ▼tDissertations Abstracts International▼g87-05B.
■790 ▼a0078
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360478▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
