Timezone: »
Causal inference requires assumptions about the data generating process, many of which are unverifiable from the data. Given that some causal assumptions might be uncertain or disputed, formal methods are needed to quantify how sensitive research conclusions are to violations of those assumptions. Although an extensive literature exists on the topic, most results are limited to specific model structures, while a general-purpose algorithmic framework for sensitivity analysis is still lacking. In this paper, we develop a formal, systematic approach to sensitivity analysis for arbitrary linear Structural Causal Models (SCMs). We start by formalizing sensitivity analysis as a constrained identification problem. We then develop an efficient, graph-based identification algorithm that exploits non-zero constraints on both directed and bidirected edges. This allows researchers to systematically derive sensitivity curves for a target causal quantity with an arbitrary set of path coefficients and error covariances as sensitivity parameters. These results can be used to display the degree to which violations of causal assumptions affect the target quantity of interest, and to judge, on scientific grounds, whether problematic degrees of violations are plausible.
Author Information
Carlos Cinelli (UCLA)
Daniel Kumor (Purdue University)
Bryant Chen (Brex)
Judea Pearl (UCLA)
Elias Bareinboim (Purdue)

Elias Bareinboim is an associate professor in the Department of Computer Science and the director of the Causal Artificial Intelligence (CausalAI) Laboratory at Columbia University. His research focuses on causal and counterfactual inference and their applications to artificial intelligence and machine learning as well as data-driven fields in the health and social sciences. His work was the first to propose a general solution to the problem of "causal data-fusion," providing practical methods for combining datasets generated under different experimental conditions and plagued with various biases. In the last years, Bareinboim has been exploring the intersection of causal inference with decision-making (including reinforcement learning) and explainability (including fairness analysis). Before joining Columbia, he was an assistant professor at Purdue University and received his Ph.D. in Computer Science from the University of California, Los Angeles. Bareinboim was named one of ``AI's 10 to Watch'' by IEEE, and is a recipient of an NSF CAREER Award, the Dan David Prize Scholarship, the 2014 AAAI Outstanding Paper Award, and the 2019 UAI Best Paper Award.
Related Events (a corresponding poster, oral, or spotlight)
-
2019 Poster: Sensitivity Analysis of Linear Structural Causal Models »
Thu. Jun 13th 01:30 -- 04:00 AM Room Pacific Ballroom #78
More from the Same Authors
-
2023 : Transportable Representations for Out-of-distribution Generalization »
Amirkasra Jalaldoust · Elias Bareinboim -
2023 Poster: Estimating Joint Treatment Effects by Combining Multiple Experiments »
Yonghan Jung · Jin Tian · Elias Bareinboim -
2022 Poster: Counterfactual Transportability: A Formal Approach »
Juan Correa · Sanghack Lee · Elias Bareinboim -
2022 Spotlight: Counterfactual Transportability: A Formal Approach »
Juan Correa · Sanghack Lee · Elias Bareinboim -
2022 Poster: Partial Counterfactual Identification from Observational and Experimental Data »
Junzhe Zhang · Jin Tian · Elias Bareinboim -
2022 Poster: On Measuring Causal Contributions via do-interventions »
Yonghan Jung · Shiva Kasiviswanathan · Jin Tian · Dominik Janzing · Patrick Bloebaum · Elias Bareinboim -
2022 Spotlight: On Measuring Causal Contributions via do-interventions »
Yonghan Jung · Shiva Kasiviswanathan · Jin Tian · Dominik Janzing · Patrick Bloebaum · Elias Bareinboim -
2022 Spotlight: Partial Counterfactual Identification from Observational and Experimental Data »
Junzhe Zhang · Jin Tian · Elias Bareinboim -
2022 : Q & A (second) »
Drago Plecko · Elias Bareinboim -
2022 : Q & A (first) »
Drago Plecko · Elias Bareinboim -
2022 Tutorial: Causal Fairness Analysis »
Elias Bareinboim · Drago Plecko -
2022 : Foundations of Causal Fairness Analysis »
Elias Bareinboim -
2021 Poster: Estimating Identifiable Causal Effects on Markov Equivalence Class through Double Machine Learning »
Yonghan Jung · Jin Tian · Elias Bareinboim -
2021 Spotlight: Estimating Identifiable Causal Effects on Markov Equivalence Class through Double Machine Learning »
Yonghan Jung · Jin Tian · Elias Bareinboim -
2020 Poster: Causal Effect Identifiability under Partial-Observability »
Sanghack Lee · Elias Bareinboim -
2020 Poster: Efficient Identification in Linear Structural Causal Models with Auxiliary Cutsets »
Daniel Kumor · Carlos Cinelli · Elias Bareinboim -
2020 Tutorial: Causal Reinforcement Learning »
Elias Bareinboim -
2019 Poster: Causal Identification under Markov Equivalence: Completeness Results »
Amin Jaber · Jiji Zhang · Elias Bareinboim -
2019 Poster: Adjustment Criteria for Generalizing Experimental Findings »
Juan Correa · Jin Tian · Elias Bareinboim -
2019 Oral: Adjustment Criteria for Generalizing Experimental Findings »
Juan Correa · Jin Tian · Elias Bareinboim -
2019 Oral: Causal Identification under Markov Equivalence: Completeness Results »
Amin Jaber · Jiji Zhang · Elias Bareinboim -
2018 Poster: Budgeted Experiment Design for Causal Structure Learning »
AmirEmad Ghassami · Saber Salehkaleybar · Negar Kiyavash · Elias Bareinboim -
2018 Oral: Budgeted Experiment Design for Causal Structure Learning »
AmirEmad Ghassami · Saber Salehkaleybar · Negar Kiyavash · Elias Bareinboim -
2017 Poster: Identification and Model Testing in Linear Structural Equation Models using Auxiliary Variables »
Bryant Chen · Daniel Kumor · Elias Bareinboim -
2017 Poster: Counterfactual Data-Fusion for Online Reinforcement Learners »
Andrew Forney · Judea Pearl · Elias Bareinboim -
2017 Talk: Counterfactual Data-Fusion for Online Reinforcement Learners »
Andrew Forney · Judea Pearl · Elias Bareinboim -
2017 Talk: Identification and Model Testing in Linear Structural Equation Models using Auxiliary Variables »
Bryant Chen · Daniel Kumor · Elias Bareinboim