Causal Identifiability

Causal identifiability focuses on determining whether a causal effect can be reliably estimated from observational data, a crucial challenge in many scientific fields. Current research emphasizes developing methods to address this challenge, particularly in the presence of unobserved confounding variables, using techniques like front-door adjustment and leveraging the power of deep generative models, including variational autoencoders and normalizing flows, to learn causal representations from data. These advancements are significant because they enable more robust causal inference in complex systems, with applications ranging from econometrics and policy evaluation to understanding the impact of algorithms and improving the reliability of counterfactual predictions in various domains.

Papers

October 3, 2023

Causal Inference with Conditional Front-Door Adjustment and Identifiable Variational Autoencoder
Ziqi Xu, Debo Cheng, Jiuyong Li, Jixue Liu, Lin Liu, Kui Yu
Causal Inference Causal Effect Estimation Partial Identifiability Variational AutoEncoder Causal Strength Front Door Adjustment Causal Identifiability

June 4, 2023

Learning Linear Causal Representations from Interventions under General Nonlinear Mixing
Simon Buchholz, Goutham Rajendran, Elan Rosenfeld, Bryon Aragam, Bernhard Schölkopf, Pradeep Ravikumar
Latent Variable Causal Representation Learning Health Intervention Global Mixing Causal Identifiability

March 2, 2023

Measuring axiomatic soundness of counterfactual image models
Miguel Monteiro, Fabio De Sousa Ribeiro, Nick Pawlowski, Daniel C. Castro, Ben Glocker
High Quality Counterfactuals Counterfactual Inference Counterfactual Image Axiomatic Approach Causal Identifiability

February 10, 2023

Causal Inference out of Control: Estimating the Steerability of Consumption
Gary Cheng, Moritz Hardt, Celestine Mendler-Dünner
External Control Causal Inference Causal Effect Consumption Data Steerable Robot Causal Identifiability

February 4, 2023

Counterfactual Identifiability of Bijective Causal Models
Arash Nasr-Esfahany, Mohammad Alizadeh, Devavrat Shah
Causal Model Counterfactual Estimation Deep Conditional Generative Causal Identifiability

January 22, 2023

Counterfactual (Non-)identifiability of Learned Structural Causal Models
Arash Nasr-Esfahany, Emre Kiciman
Causal Structure Structural Causal Model Counterfactual Inference Structural Causal Causal Identifiability Deep Structural Causal Non Identifiability

August 15, 2022

Anticipating Performativity by Predicting from Predictions
Celestine Mendler-Dünner, Frances Ding, Yixin Wang
Human Prediction Causal Effect Causal Relationship Causal Inference Task Performative Power Causal Identifiability

February 7, 2022

CITRIS: Causal Identifiability from Temporal Intervened Sequences
Phillip Lippe, Sara Magliacane, Sindy Löwe, Yuki M. Asano, Taco Cohen, Efstratios Gavves
Supervised Autoencoder Causal Representation Learning Temporal Sequence Causal Hierarchy Clover Sowing Causal Identifiability