III: Medium: Collaborative Research: A Unified and Declarative Approach to Causal Analysis for Big Data

III：媒介：协作研究：大数据因果分析的统一声明式方法

• 批准号: 1703431
• 负责人: Sudeepa Roy
• 金额: $ 40.8万
• 依托单位: Duke University
• 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 2022-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1703431&HistoricalAwards=false
• 关键词: III; Medium; Collaborative; Research; Unified

Observational data is available today in multi-relational form, often extracted from various sources, and stored in multiple flat and interrelated tables. Standard statistical methods for conducting causal inference on observational data assume a very simple data model: a single table with independent units. This research has the potential to significantly impact application domains where differentiating causality from correlation is essential, e.g., education policy and cancer genomics. The HUME project develops techniques for efficient causal analysis using a declarative approach, over complex views, and over large datasets that are integrated from disparate data sources. HUME uses a SQL-like language and is integrated with a relational database system.The project develops techniques for defining arbitrarily complex units, treatments, outcomes, and covariates, by combining joins, data mapping, and aggregates across multiple tables, and uses a causal network to choose a good set of covariates for causal inference. The first part of the project develops scalable techniques for sub-classification and matching for large data sets obtained by declaratively integrating multiple data sources. The second part of the project develops scalable methods for discovering causal relationships among the attributes in the views by constraint-based, search-based, and hybrid discovery processes. Finally, the third part of the project investigates interferences among units arising from the complex views by designing normal forms and automatic inference of underlying assumptions exploiting techniques from database theory.

如今，观测数据以多关系形式提供，通常从各种来源提取，并存储在多个平面且相互关联的表中。 对观测数据进行因果推断的标准统计方法假设一个非常简单的数据模型：具有独立单元的单个表。 这项研究有可能对区分因果关系和相关性至关重要的应用领域产生重大影响，例如教育政策和癌症基因组学。 HUME 项目开发了使用声明性方法、针对复杂视图以及从不同数据源集成的大型数据集进行有效因果分析的技术。 HUME 使用类似 SQL 的语言，并与关系数据库系统集成。该项目开发了通过组合联接、数据映射和跨多个表的聚合来定义任意复杂的单元、处理、结果和协变量的技术，并使用因果关系网络选择一组好的协变量进行因果推理。 该项目的第一部分开发可扩展的技术，用于对通过声明式集成多个数据源获得的大型数据集进行子分类和匹配。 该项目的第二部分开发可扩展的方法，通过基于约束、基于搜索和混合的发现过程来发现视图中属性之间的因果关系。最后，该项目的第三部分通过设计范式和利用数据库理论技术自动推断基本假设来研究由于复杂视图而产生的单元之间的干扰。

项目成果

FLAME: A Fast Large-scale Almost Matching Exactly Approach to Causal Inference

FLAME：一种快速、大规模、几乎完全匹配的因果推理方法

• 发表时间: 2017-07-19
• 期刊: J. Mach. Learn. Res.
• 作者: Sudeepa Roy;C. Rudin;A. Volfovsky;Tianyu Wang
• 通讯作者: Tianyu Wang

Adaptive Hyper-box Matching for Interpretable Individualized Treatment Effect Estimation

用于可解释的个体化治疗效果估计的自适应超框匹配

• 发表时间: 2020-01
• 期刊: Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI
• 作者: Morucci, Marco;Orlandi, Vittorio;Roy, Sudeepa;Rudin, Cynthia;Volfovsky, Alexander
• 通讯作者: Volfovsky, Alexander

iQCAR: inter-Query Contention Analyzer for Data Analytics Frameworks

iQCAR：数据分析框架的查询间争用分析器

• DOI: 10.1145/3299869.3319904
• 发表时间: 2019-06
• 期刊: ACM SIGMOD International Conference on Management of Data (SIGMOD
• 作者: Kalmegh P;Babu S;Roy S
• 通讯作者: Roy S

Almost-Matching-Exactly for Treatment Effect Estimation under Network Interference

网络干扰下的几乎匹配-精确治疗效果估计

• 发表时间: 2020-01
• 期刊: Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics
• 作者: Awan, Usaid;Morucci, Marco;Orlandi, Vittorio;Roy, Sudeepa;Rudin, Cynthia;Volfovsky, Alexander
• 通讯作者: Volfovsky, Alexander

Interpretable Almost Matching Exactly With Instrumental Variables

可解释的几乎与工具变量完全匹配

• 发表时间: 2019-01
• 期刊: Conference on Uncertainty in Artificial Intelligence (UAI
• 作者: Awan, M Usaid;Liu, Yameng;Morucci, Marco;Roy, Sudeepa;Rudin, Cynthia;Volfovsky, Alexander
• 通讯作者: Volfovsky, Alexander