서브메뉴
검색
On Some Topics in Statistical Learning Cluster-Aware Lasso & Others
On Some Topics in Statistical Learning Cluster-Aware Lasso & Others
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202104734
- ISBN
- 9798290649665
- DDC
- 519.77
- 서명/저자
- On Some Topics in Statistical Learning Cluster-Aware Lasso & Others
- 발행사항
- [Sl] : Stanford University, 2023
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2023
- 형태사항
- 136 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-03, Section: B.
- 주기사항
- Advisor: Tibshirani, Robert.
- 학위논문주기
- Thesis (Ph.D.)--Stanford University, 2023.
- 초록/해제
- 요약In this thesis, we visit four topics in statistical learning:• Cluster-Aware Lasso. An adaptation of lasso dealing with correlated features in supervised learning, which uses a hierarchical clustering-based approach to adaptively select clusters of features.• Confidence Intervals for Generalisation Error in Random Forests. An effort to extend the out-of-bag error point estimate in random forests to a confidence interval with appropriate coverage for generalisation error.• Prediction of Gestational age using metabolite data. Approaches to a specific problem type in supervised learning in which multiple observations are made at different time points for a given unit and the response is a time to or from a fixed event.• Statistical Summaries of unlabelled evolutionary trees. Techniques for summarising samples and distributions on a class of tree structures which are ranked and unlabelled.While these four topics are somewhat disparate, they all fall under the overall umbrella of statistical learning, and are relevant in dealing with new classes of data, especially in biological or high-dimensional contexts.As an amusing connect, three of the four chapters deal with tree-like structures, though very different ones. The cluster-aware lasso deals strongly with hierarchical clustering dendrograms on the features in supervised learning. Random forests use decision trees as the base learner. In our final chapter, we deal with trees in which time is an axis and we track the evolutionary history of a set of objects.
- 일반주제명
- Integer programming
- 일반주제명
- Signal to noise ratio
- 일반주제명
- Gestational age
- 일반주제명
- Statistics
- 기타저자
- Stanford University.
- 기본자료저록
- Dissertations Abstracts International. 87-03B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2023 us c eng d■001000017358664
■00520260202104734
■006m o d
■007cr#unu||||||||
■020 ▼a9798290649665
■035 ▼a(MiAaPQ)AAI32149628
■035 ▼a(MiAaPQ)Stanfordcg748xm7766
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a519.77
■1001 ▼aSamyak, Rajanala.
■24510▼aOn Some Topics in Statistical Learning Cluster-Aware Lasso & Others
■260 ▼a[Sl]▼bStanford University▼c2023
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2023
■300 ▼a136 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-03, Section: B.
■500 ▼aAdvisor: Tibshirani, Robert.
■5021 ▼aThesis (Ph.D.)--Stanford University, 2023.
■520 ▼aIn this thesis, we visit four topics in statistical learning:• Cluster-Aware Lasso. An adaptation of lasso dealing with correlated features in supervised learning, which uses a hierarchical clustering-based approach to adaptively select clusters of features.• Confidence Intervals for Generalisation Error in Random Forests. An effort to extend the out-of-bag error point estimate in random forests to a confidence interval with appropriate coverage for generalisation error.• Prediction of Gestational age using metabolite data. Approaches to a specific problem type in supervised learning in which multiple observations are made at different time points for a given unit and the response is a time to or from a fixed event.• Statistical Summaries of unlabelled evolutionary trees. Techniques for summarising samples and distributions on a class of tree structures which are ranked and unlabelled.While these four topics are somewhat disparate, they all fall under the overall umbrella of statistical learning, and are relevant in dealing with new classes of data, especially in biological or high-dimensional contexts.As an amusing connect, three of the four chapters deal with tree-like structures, though very different ones. The cluster-aware lasso deals strongly with hierarchical clustering dendrograms on the features in supervised learning. Random forests use decision trees as the base learner. In our final chapter, we deal with trees in which time is an axis and we track the evolutionary history of a set of objects.
■590 ▼aSchool code: 0212.
■650 4▼aInteger programming
■650 4▼aSignal to noise ratio
■650 4▼aSevere acute respiratory syndrome coronavirus 2
■650 4▼aGestational age
■650 4▼aStatistics
■653 ▼aStatistical learning
■653 ▼aSupervised learning
■690 ▼a0463
■71020▼aStanford University.
■7730 ▼tDissertations Abstracts International▼g87-03B.
■790 ▼a0212
■791 ▼aPh.D.
■792 ▼a2023
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17358664▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
