Deep Structural Causal Models for Tractable Counterfactual Inference

TL;DR

This paper introduces a deep structural causal modeling framework using normalising flows and variational inference, enabling tractable counterfactual inference in complex data like images and medical scans.

Contribution

It presents a novel deep SCM framework that allows for efficient counterfactual inference, filling a gap in existing deep causal learning methods.

Findings

Successfully trained deep SCMs on synthetic and real-world datasets.

Capable of answering association, intervention, and counterfactual questions.

Provides a new tool for causal analysis in imaging applications.

Abstract

We formulate a general framework for building structural causal models (SCMs) with deep learning components. The proposed approach employs normalising flows and variational inference to enable tractable inference of exogenous noise variables - a crucial step for counterfactual inference that is missing from existing deep causal learning methods. Our framework is validated on a synthetic dataset built on MNIST as well as on a real-world medical dataset of brain MRI scans. Our experimental results indicate that we can successfully train deep SCMs that are capable of all three levels of Pearl's ladder of causation: association, intervention, and counterfactuals, giving rise to a powerful new approach for answering causal questions in imaging applications and beyond. The code for all our experiments is available at https://github.com/biomedia-mira/deepscm.