서브메뉴
검색
Algorithm Dynamics in Modern Statistical Learning: Asymptotics, Universality, and Implicit Regularization
Algorithm Dynamics in Modern Statistical Learning: Asymptotics, Universality, and Implicit Regularization
Detailed Information
- Material Type
- 단행본
- 0017160537
- Date and Time of Latest Transaction
- 20250211151036
- ISBN
- 9798383482179
- DDC
- 310
- Author
- Wang, Tianhao.
- Title/Author
- Algorithm Dynamics in Modern Statistical Learning: Asymptotics, Universality, and Implicit Regularization
- Publish Info
- [Sl] : Yale University, 2024
- Publish Info
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- Material Info
- 349 p
- General Note
- Source: Dissertations Abstracts International, Volume: 86-01, Section: B.
- General Note
- Advisor: Fan, Zhou.
- 학위논문주기
- Thesis (Ph.D.)--Yale University, 2024.
- Abstracts/Etc
- 요약Understanding the dynamics of algorithms is crucial for characterizing the behavior of trained models in modern statistical learning. This thesis presents a few recent results on theoretical analyses of the dynamics of two classes of algorithms: Approximate Message Passing (AMP) algorithms and Stochastic Gradient Descent (SGD). For AMP algorithms, the focus is to derive the precise asymptotic distributional characterization of the iterates, known as the ``state evolution'' which summarizes the dynamics of AMP iterates, and to understand the universality of such characterization with respect to the underlying data distribution. For SGD, the goal is to perform trajectory analysis to understand its implicit regularization, a key property believed to be essential for the generalization of modern deep learning models. The results presented here provide unified frameworks for analyzing and understanding the dynamics of these algorithms, and can be potentially extended to other algorithms.
- Subject Added Entry-Topical Term
- Statistics
- Index Term-Uncontrolled
- Approximate message passing
- Index Term-Uncontrolled
- Implicit regularization
- Index Term-Uncontrolled
- Stochastic gradient descent
- Index Term-Uncontrolled
- Universality
- Added Entry-Corporate Name
- Yale University Statistics and Data Science
- Host Item Entry
- Dissertations Abstracts International. 86-01B.
- Electronic Location and Access
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017160537
■00520250211151036
■006m o d
■007cr#unu||||||||
■020 ▼a9798383482179
■035 ▼a(MiAaPQ)AAI30997623
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a310
■1001 ▼aWang, Tianhao.
■24510▼aAlgorithm Dynamics in Modern Statistical Learning: Asymptotics, Universality, and Implicit Regularization
■260 ▼a[Sl]▼bYale University▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a349 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-01, Section: B.
■500 ▼aAdvisor: Fan, Zhou.
■5021 ▼aThesis (Ph.D.)--Yale University, 2024.
■520 ▼aUnderstanding the dynamics of algorithms is crucial for characterizing the behavior of trained models in modern statistical learning. This thesis presents a few recent results on theoretical analyses of the dynamics of two classes of algorithms: Approximate Message Passing (AMP) algorithms and Stochastic Gradient Descent (SGD). For AMP algorithms, the focus is to derive the precise asymptotic distributional characterization of the iterates, known as the ``state evolution'' which summarizes the dynamics of AMP iterates, and to understand the universality of such characterization with respect to the underlying data distribution. For SGD, the goal is to perform trajectory analysis to understand its implicit regularization, a key property believed to be essential for the generalization of modern deep learning models. The results presented here provide unified frameworks for analyzing and understanding the dynamics of these algorithms, and can be potentially extended to other algorithms.
■590 ▼aSchool code: 0265.
■650 4▼aStatistics
■653 ▼aApproximate message passing
■653 ▼aImplicit regularization
■653 ▼aStochastic gradient descent
■653 ▼aUniversality
■690 ▼a0463
■71020▼aYale University▼bStatistics and Data Science.
■7730 ▼tDissertations Abstracts International▼g86-01B.
■790 ▼a0265
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17160537▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
Preview
Export
ChatGPT Discussion
AI Recommended Related Books
Detail Info.
- Reservation
- Not Exist
- My Folder
- First Request
- Non-Book Loan Application
- Nighttime Book Loan Application
Available after logging in.


