## 概述

• 日期: 2022年04月02日
• 位置: 中国科学院计算技术研究所四层报告厅

## 首页

2022“因果推断与机器学习”研讨会组委会

## 日程

4月1日 14:00-22:00 报到注册
4月2日 08:50-09:10 开幕式、致辞 郭嘉丰
09:10-09:40 林华珍（西南财经大学） Robust and efficient estimation for treatment effect in causal inference 陈薇
09:40-10:10 崔鹏（清华大学） 因果启发的稳定学习
10:10-10:40 林伟（北京大学） Deconfounding with the Blessing of Dimensionality
10:40-11:00 茶歇
11:00-11:30 苗旺（北京大学） 因果推断，观察性研究和诺贝尔经济学奖 邹长亮
11:30-12:00 丁锐（微软亚洲研究院） Supervised Causal Learning: A New Frontier of Causal Discovery
12:00-12:30 王磊（南开大学） Generalized regression estimators for average treatment effect with multicollinearity in high-dimensional covariates
12:30-14:00 午餐
14:00-14:30 陈卫（微软亚洲研究院） Combinatorial Causal Bandit 苗旺
14:30-15:00 周岭（西南财经大学） Confederated learning and Inference
15:00-15:30 张政（中国人民大学） Nonparametric Estimation of Continuous Treatment Effect with Measurement Error
15:30-15:50 茶歇
15:50-16:20 刘林（上海交通大学） A novel stable higher-order influence function estimators for doubly-robust functionals 陈卫
16:20-16:50 李伟（中国人民大学） Estimation and inference for high-dimensional nonparametric additive instrumental-variables regression
16:50-17:20 刘畅（微软亚洲研究院） Improving out-of-Distribution Performance of Machine Learning Models from a Causal Perspective

## 报告嘉宾

#### 李伟（中国人民大学）

high-dimensional linear instrumental-variables model has been considered in the literature due to its simplicity albeit the true relationship may be nonlinear. We propose a more data-driven approach by considering nonparametric additive models between the instrumental variables and the treatments while keeping the linear model assumption between the treatments and the outcome so that the coefficients therein can directly bear causal interpretation. We provide a two-stage framework for estimation and inference under this more general setup. The group lasso regularization is first employed to select optimal instruments for the high-dimensional nonparametric additive model, and then the outcome variable is regressed on the fitted values from the nonparametric additive model to identify and estimate important treatment effects. We provide non-asymptotic analysis for the estimation error of the proposed estimator. A debiased procedure is further employed to establish valid inference. Extensive numerical experiments show that our proposed method can rival or outperform existing approaches in the literature. We finally analyze the mouse obesity data with the proposed method and discuss new discoveries.