III: Small: Collaborative Research: Comprehensive Heterogeneous Response Regression from Complex Data

III：小：协作研究：复杂数据的综合异质响应回归

• 批准号: 1718798
• 负责人: Kun Chen
• 金额: $ 25万
• 依托单位: University of Connecticut
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-09-01 至 2021-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1718798&HistoricalAwards=false
• 关键词: III; Small; Collaborative; Research; Comprehensive

Predictive modeling is one fundamental problem in supervised machine learning. Traditional predictive modeling approaches typically built one predictor for each prediction task. However, in many real world problems, one needs to build predictors for multiple inter-correlated tasks simultaneously. For example, in real-time anesthesia decision making, the anesthesia drugs will have impact on multiple indicators of an anesthesia patient, such as anesthesia depth, blood pressures, heart rates, etc. The anesthesiologist needs to consider all those different aspects as well as their intrinsic dependence before s/he can make the decision. The goal of this project is to conduct systematic research on heterogeneous response regression, which builds multiple regression models for heterogeneous responses as well as exploring the relationship among them.Specifically, the project's heterogeneous response regression framework is based on a tailored latent factor model that captures the relationship among different responses in a low-dimensional space. The response heterogeneity will be captured by utilizing the exponential family distributions and beyond. The model parameters can be learned through regularized maximum likelihood estimation via a majorization-minimization procedure. Moreover, a distributed solution based on the alternating direction method of multipliers can be derived to enhance the scalability. The project will also make the framework more flexible by (1) adding individualized effect terms to capture sample/feature heterogeneity and induce task dependency; and (2) leveraging a functional model to account for time-varying predictors and to make the prediction time-sensitive. The project will demonstrate the effectiveness of its method in disease risk prediction and financial stability modeling.

预测建模是监督机器学习中的基本问题之一。传统的预测建模方法通常为每个预测任务构建一个预测器。然而，在许多现实世界的问题中，需要同时为多个相互关联的任务构建预测器。例如，在实时麻醉决策中，麻醉药物会影响麻醉患者的多个指标，如麻醉深度、血压、心率等。麻醉师需要考虑所有这些不同的方面以及它们的内在依赖性才能做出决定。该项目的目标是对异质响应回归进行系统研究，为异质响应建立多个回归模型并探索它们之间的关系。具体来说，该项目的异质响应回归框架基于定制的潜在因子模型，该模型捕获低维空间中不同响应之间的关系。响应异质性将通过利用指数族分布及其他分布来捕获。模型参数可以通过主化-最小化过程的正则化最大似然估计来学习。此外，可以导出基于乘法器交替方向方法的分布式解决方案以增强可扩展性。该项目还将通过以下方式使框架更加灵活：（1）添加个性化效果项以捕获样本/特征异质性并诱导任务依赖性； (2) 利用函数模型来考虑随时间变化的预测变量并使预测对时间敏感。该项目将展示其方法在疾病风险预测和金融稳定建模方面的有效性。

项目成果

Efficient interaction selection for clustered data via stagewise generalized estimating equations

通过阶段广义估计方程对聚类数据进行有效的交互选择

• DOI: 10.1002/sim.8574
• 发表时间: 2020
• 期刊: Statistics in Medicine
• 影响因子: 2
• 作者: Vaughan, Gregory;Aseltine, Robert;Chen, Kun;Yan, Jun
• 通讯作者: Yan, Jun

Robust finite mixture regression for heterogeneous targets

• DOI: 10.1007/s10618-018-0564-z
• 发表时间: 2018-04
• 期刊: Data Mining and Knowledge Discovery
• 影响因子: 4.8
• 作者: Jian Liang;Kun Chen;Ming Lin;Changshui Zhang;Fei Wang

Integrative survival analysis with uncertain event times in application to a suicide risk study

不确定事件时间的综合生存分析在自杀风险研究中的应用

• DOI: 10.1214/19-aoas1287
• 发表时间: 2020
• 期刊: The Annals of Applied Statistics
• 作者: Wang, Wenjie;Aseltine, Robert;Chen, Kun;Yan, Jun
• 通讯作者: Yan, Jun

Leveraging mixed and incomplete outcomes via reduced-rank modeling

• DOI: 10.1016/j.jmva.2018.04.011
• 发表时间: 2018-09
• 期刊: J. Multivar. Anal.
• 作者: Chongliang Luo;Jian Liang;Gen Li;Fei Wang;Changshui Zhang;D. Dey;Kun Chen

Multivariate Functional Regression Via Nested Reduced-Rank Regularization

• DOI: 10.1080/10618600.2021.1960850
• 发表时间: 2020-03
• 期刊: Journal of Computational and Graphical Statistics
• 影响因子: 2.4
• 作者: Xiaokang Liu;Shujie Ma;Kun Chen