서브메뉴
검색
Estimation and Inference in Causal Models and Multi-Modal Knowledge Graph Integration
Estimation and Inference in Causal Models and Multi-Modal Knowledge Graph Integration
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202103524
- ISBN
- 9798280719194
- DDC
- 310
- 저자명
- Chen, Qizhao.
- 서명/저자
- Estimation and Inference in Causal Models and Multi-Modal Knowledge Graph Integration
- 발행사항
- [Sl] : Harvard University, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 213 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
- 주기사항
- Advisor: Austern, Morgane;Cai, Tianxi.
- 학위논문주기
- Thesis (Ph.D.)--Harvard University, 2025.
- 초록/해제
- 요약This dissertation examines the estimation and inference of causal parameters in two different frameworks: the generalized method of moments framework and the dynamic optimal treatment regime framework. For generalized method of moments framework: Chapter 1 shows that when the auxiliary estimators satisfy a leave-one-out stability condition, debiased machine learning can achieve root-n consistency and asymptotic normality without requiring sample splitting or cross-fitting. This enables more efficient sample reuse, especially in moderate-sample regimesFor dynamic optimal treatment regimes: Chapter 2 analyzes the statistical properties of a softmax approximation to the optimal policy. It demonstrates that under a suitable growing scheme of the temperature parameter, this softmax approach yields valid inference for the value and structural parameters associated with the true optimal regime.Apart from causal parameter estimation and inference, this dissertation also advances causal understanding by integrating biomedical knowledge to uncover biological drivers of clinical outcomes and to inform clinical decision-making:Biomedical knowledge graph integration: Chapter 3 constructs a heterogeneous, multi-modal knowledge graph using a Relational Graph Convolutional Network (R-GCN) to embed clinical and biological entities, including disease, drugs, genes, and single nucleotide polymorphisms (SNPs), into a unified representation that supports various link prediction tasks and downstream applications.
- 일반주제명
- Statistics
- 일반주제명
- Computer engineering
- 일반주제명
- Computer science
- 키워드
- Machine learning
- 키워드
- Softmax approach
- 기타저자
- Harvard University Statistics
- 기본자료저록
- Dissertations Abstracts International. 86-12B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017357527
■00520260202103524
■006m o d
■007cr#unu||||||||
■020 ▼a9798280719194
■035 ▼a(MiAaPQ)AAI32039170
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a310
■1001 ▼aChen, Qizhao.▼0(orcid)0009-0001-1329-3303
■24510▼aEstimation and Inference in Causal Models and Multi-Modal Knowledge Graph Integration
■260 ▼a[Sl]▼bHarvard University▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a213 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-12, Section: B.
■500 ▼aAdvisor: Austern, Morgane;Cai, Tianxi.
■5021 ▼aThesis (Ph.D.)--Harvard University, 2025.
■520 ▼aThis dissertation examines the estimation and inference of causal parameters in two different frameworks: the generalized method of moments framework and the dynamic optimal treatment regime framework. For generalized method of moments framework: Chapter 1 shows that when the auxiliary estimators satisfy a leave-one-out stability condition, debiased machine learning can achieve root-n consistency and asymptotic normality without requiring sample splitting or cross-fitting. This enables more efficient sample reuse, especially in moderate-sample regimesFor dynamic optimal treatment regimes: Chapter 2 analyzes the statistical properties of a softmax approximation to the optimal policy. It demonstrates that under a suitable growing scheme of the temperature parameter, this softmax approach yields valid inference for the value and structural parameters associated with the true optimal regime.Apart from causal parameter estimation and inference, this dissertation also advances causal understanding by integrating biomedical knowledge to uncover biological drivers of clinical outcomes and to inform clinical decision-making:Biomedical knowledge graph integration: Chapter 3 constructs a heterogeneous, multi-modal knowledge graph using a Relational Graph Convolutional Network (R-GCN) to embed clinical and biological entities, including disease, drugs, genes, and single nucleotide polymorphisms (SNPs), into a unified representation that supports various link prediction tasks and downstream applications.
■590 ▼aSchool code: 0084.
■650 4▼aStatistics
■650 4▼aComputer engineering
■650 4▼aComputer science
■653 ▼aCausal parameters
■653 ▼aMachine learning
■653 ▼aSoftmax approach
■653 ▼aRelational Graph Convolutional Network
■653 ▼aSingle nucleotide polymorphisms
■690 ▼a0463
■690 ▼a0984
■690 ▼a0464
■690 ▼a0800
■71020▼aHarvard University▼bStatistics.
■7730 ▼tDissertations Abstracts International▼g86-12B.
■790 ▼a0084
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17357527▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
