OCIS · Fall 2022¶
- 共 13 场 · 11 篇精读
本季导览¶
自动生成：归纳本季主线与值得先看的几场，不打分、不排名。
这一季的 13 场报告可归纳为四条主线：因果推断中的统一框架与优化视角（Zubizarreta、Lei、Singh）、因果发现与表示学习（Lam、Zhang、Huang）、实验设计与网络因果（Rajkumar、Josse）、以及半参数效率与自动推断（Singh、Syrgkanis、Miratrix、Li、Janzing）。其中，Zubizarreta 和 Lei 分别从匹配/加权和面板数据 TWFE 出发，试图用数学规划或双稳健方法统一经典调整手段；Singh 的两场报告则聚焦于数据污染（测量误差、缺失、差分隐私）下的半参数推断，与 Syrgkanis 的自动去偏机器学习形成互补——前者处理输入数据质量，后者自动化去偏项的构造。
最突出的主线是“半参数效率与自动推断”，涉及 Singh（两场）、Syrgkanis、Miratrix 和 Li。Singh 的第一场报告将测量误差、缺失、离散化和差分隐私统一视为“数据污染”，提出基于低秩假设的矩阵补全+去偏估计量，并给出有限样本高斯逼近率；Syrgkanis 则从另一方向切入，用递归 Riesz 表示子自动构造动态治疗效应等嵌套泛函的去偏估计量，无需手动推导影响函数。Miratrix 的 DiD+匹配工作可视为这条线的特例——它分析“先匹配再 DiD”的偏差-偏差权衡，本质上是在平行趋势假设不成立时，用匹配作为去偏手段。Li 的贝叶斯因果推断教程则提供了与频率学派半参数方法平行的贝叶斯视角，强调缺失数据插补框架下的不确定性量化。
第二条主线是“因果发现与表示学习”，由 Lam、Zhang 和 Huang 构成。Lam 推广了基于稀疏排列的 DAG 学习算法（GRaSP），在比忠实性更弱的假设下实现一致估计；Zhang 系统介绍了因果表示学习，从低层观测变量推断高层潜变量及其因果结构，涵盖非线性 ICA、独立变化机制等工具；Huang 则聚焦于潜变量层次结构发现，利用秩约束（Tetrad 条件）从观测叶节点恢复潜变量树或更一般的图。三者共享“从观测数据恢复潜在因果结构”的目标，但分别从排列搜索、非线性解耦和秩约束切入。
第三条主线是“实验设计与网络因果”，以 Rajkumar 和 Josse 为代表。Rajkumar 利用 LinkedIn 的大规模随机实验（推荐算法 A/B 测试）检验弱连接对求职的因果效应，处理了网络内生性和边级样本选择偏差；Josse 则结合 RCT 与观察性数据（脑外伤治疗），讨论如何利用不完整数据（缺失、混杂）进行因果推断。两者都涉及真实世界实验的复杂性——前者是平台实验中的干扰与多维度处理，后者是数据缺失与外部有效性。
第四条主线是“统一框架与优化视角”，包括 Zubizarreta、Lei 和 Janzing。Zubizarreta 将匹配、回归和加权统一为数学规划问题，强调有限样本下对随机化特征的精确逼近；Lei 提出双稳健 TWFE 估计量，通过单位特定权重修正异质性处理效应下的负权重问题；Janzing 则为根因分析（RCA）提供了基于因果 DAG 的定量框架，区别于传统的因果效应估计。三者都试图用优化或代数结构统一看似分散的方法。
推荐观看路径：若想快速了解半参数效率与自动推断，可先看 Singh 的第一场（数据污染下的去偏推断）和 Syrgkanis（AutoDML），再以 Miratrix（DiD+匹配的偏差分析）作为应用案例。若对因果发现感兴趣，建议先看 Lam（GRaSP 算法）打底，再进入 Zhang（因果表示学习）和 Huang（潜变量层次结构）的进阶内容。若关注实验设计与网络因果，Rajkumar 的 LinkedIn 实验是理解平台实验设计的良好入口，Josse 的脑外伤研究则展示了 RCT+观察数据的融合。若想理解因果推断的底层统一视角，Zubizarreta 的数学规划框架和 Lei 的双稳健 TWFE 是互补的起点。
报告列表¶
Bridging Matching, Regression, and Weighting as Mathematical Programs for Causal Inference¶
讲者: Jose Zubizarreta · 讨论人: Mike Baiocchi · 2022-12-06
链接：视频
摘要
A fundamental principle in the design of observational studies is to approximate the randomized experiment that would have been conducted under controlled circumstances. Across the health and social sciences, statistical methods for covariate adjustment are used in pursuit of this principle. Basic methods are matching, regression, and weighting. In this talk, we will examine the connections between these methods through their underlying mathematical programs. We will study their strengths and we…Greedy Relaxations of the Sparsest Permutation Algorithm¶
讲者: Wayne Lam · 讨论人: Alex Markham · 2022-11-29
链接：视频 · 幻灯片
摘要
There has been an increasing interest in methods that exploit permutation reasoning to search for directed acyclic causal models, including the "Ordering Search" of Teyssier and Kohler (2005), and GSP of Solus, Wang and Uhler (2021). We extend the methods of the latter by a permutation-based operation, tuck, and develop a class of algorithms, namely GRaSP, that are efficient and pointwise consistent under increasingly weaker assumptions than faithfulness. The most relaxed form of GRaSP outperfor…Causal inference for brain trauma: leveraging incomplete observational data and RCT （暂无精读）¶
讲者: Julie Josse · 讨论人: Elizabeth Stuart · 2022-11-22
链接：视频 · 幻灯片 · arXiv
摘要
The simultaneous availability of observational and experimental data for the same medical question about the effect of a treatment is an opportunity to combine their strengths and address their weaknesses. In this presentation, I will illustrate the methodological challenges we faced in answering a medical question about the effect of tranexamic acid administration on mortality in patients with traumatic brain injury in the context of critical care management. First, we had access to a large Fre…A causal test of the strength of weak ties¶
讲者: Karthik Rajkumar · 讨论人: Dean Eckles · 2022-11-15
链接：视频 · 幻灯片
摘要
We analyzed data from multiple large-scale randomized experiments on LinkedIn’s People You May Know algorithm, which recommends new connections to LinkedIn members, to test the extent to which weak ties increased job mobility in the world’s largest professional social network. The experiments randomly varied the prevalence of weak ties in the networks of over 20 million people over a 5-year period, during which 2 billion new ties and 600,000 new jobs were created. The results provided experiment…A devil’s bargain? Repairing a Difference in Differences parallel trends assumption with an initial matching step¶
讲者: Luke Miratrix · 讨论人: Laura Hatfield · 2022-11-08
链接：视频 · 幻灯片
摘要
The Difference in Difference (DiD) estimator is a popular estimator built on the "parallel trends" assumption that the treatment group, absent treatment, would change "similarly" to the control group over time. To increase the plausibility of this assumption, a natural idea is to match treated and control units prior to a DiD analysis. In this paper, we characterize the bias of matching under a class of linear structural models with both observed and unobserved confounders that have time varying…Methodological advances in causal representation learning¶
讲者: Kun Zhang · 讨论人: Victor Veitch · 2022-11-01
链接：视频 · 幻灯片
摘要
Causal representation learning aims to reveal the underlying high-level hidden causal variables and their relations. It can be seen as a special case of causal discovery, whose goal is to recover the underlying causal structure or causal model from observational data. The modularity property of a causal system implies properties of minimal changes and independent changes of causal representations, and how such properties make it possible to recover the underlying causal representations from obse…Causal Inference with Corrupted Data: Measurement Error, Missing Values, Discretization, and Differential Privacy¶
讲者: Rahul Singh, Talk 1 · 2022-10-25
链接：视频 · 幻灯片
- Talk 1 Title: Causal Inference with Corrupted Data: Measurement Error, Missing Values, Discretization, and Differential Privacy - Talk 1 Abstract: The 2020 US Census will be published with differential privacy, implemented by injecting synthetic noise into the data. Controversy has ensued, with debates that center on the painful trade-off between the privacy of respondents and the precision of economic analysis. Is this trade-off inevitable? To answer this question, we formulate a semiparametric model of causal inference with high dimensional data that may be noisy, missing, discretized, or privatized. We propose a new end-to-end procedure for data cleaning, estimation, and inference with data cleaning-adjusted confidence intervals. We prove consistency, Gaussian approximation, and semiparametric efficiency by finite sample arguments. The rate of Gaussian approximation is n−1/2 for semiparametric estimands such as average treatment effect, and it degrades gracefully for nonparametric estimands such as heterogeneous treatment effect. Our key assumption is that the true covariates are approximately low rank, which we interpret as approximate repeated measurements and validate in the Census. In our analysis, we provide nonasymptotic theoretical contributions to matrix completion, statistical learning, and semiparametric statistics. We verify the coverage of the data cleaning-adjusted confidence intervals in simulations. Finally, we conduct a semi-synthetic exercise calibrated to privacy levels mandated for the 2020 US Census. （暂无精读）¶
讲者: Rahul Singh · 2022-10-25
链接：视频 · 幻灯片 · arXiv
Latent Hierarchical Causal Structure Discovery with Rank Constraints¶
讲者: Biwei Huang · 讨论人: Erich Kummerfeld · 2022-10-18
链接：视频 · 幻灯片 · arXiv
A tutorial on Bayesian causal inference¶
讲者: Fan Li · 2022-10-11
链接：视频 · 幻灯片
摘要
This paper provides a critical review of the Bayesian perspective of causal inference based on the potential outcomes framework. We review the causal estimands, identification assumptions, and general structure of Bayesian inference of causal effects. We highlight issues that are unique to Bayesian causal inference, including the role of the propensity score, definition of identifiability, and choice of priors in both low and high dimensional regimes. We point out the central role of covariate o…Double-Robust Two-Way-Fixed-Effects Regression For Panel Data¶
讲者: Lihua Lei · 讨论人: Jeffrey Wooldridge · 2022-10-04
链接：视频 · 幻灯片 · arXiv
摘要
We propose a new estimator for the average causal effects of a binary treatment with panel data in settings with general treatment patterns. Our approach augments the two-way-fixed-effects specification with the unit-specific weights that arise from a model for the assignment mechanism. We show how to construct these weights in various settings, including situations where units opt into the treatment sequentially. The resulting estimator converges to an average (over units and time) treatment ef…Automatic Debiased Machine Learning for Dynamic Treatment Effects and General Nested Functionals¶
讲者: Vasilis Syrgkanis · 讨论人: Eric Tchetgen Tchetgen · 2022-09-27
链接：视频 · arXiv
摘要
We extend the idea of automated debiased machine learning to the dynamic treatment regime and more generally to nested functionals. We show that the multiply robust formula for the dynamic treatment regime with discrete treatments can be re-stated in terms of a recursive Riesz representer characterization of nested mean regressions. We then apply a recursive Riesz representer estimation learning algorithm that estimates de-biasing corrections without the need to characterize how the correction t…- F ormal framework for quantitative Root Cause Analysis¶
讲者: Dominik Janzing · 讨论人: Niklas Pfister · 2022-09-20
链接：视频 · 幻灯片 · arXiv
摘要
Asking for the “root cause(s)” of a singular event is at the heart of human attempts to understand what happened. Nevertheless, we were not able to find a satisfactory formalization of “Root Cause Analysis (RCA)” for our business. We have therefore proposed a framework for RCA of anomalies [ 1Maintained by 陈星宇 · Homepage · Source