[2011.02268] Causal Autoregressive Flows

Two apparently unrelated fields -- normalizing flows and causality -- have recently received considerable attention in the machine learning community. In this work, we highlight an intrinsic correspondence between a simple family of autoregressive normalizing flows and identifiable causal models. We exploit the fact that autoregressive flow architectures define an ordering over variables, analogous to a causal ordering, to show that they are well-suited to performing a range of causal inference tasks, ranging from causal discovery to making interventional and counterfactual predictions. First, we show that causal models derived from both affine and additive autoregressive flows with fixed orderings over variables are identifiable, i.e. the true direction of causal influence can be recovered. This provides a generalization of the additive noise model well-known in causal discovery. Second, we derive a bivariate measure of causal direction based on likelihood ratios, leveraging the fact that flow models can estimate normalized log-densities of data. Third, we demonstrate that flows naturally allow for direct evaluation of both interventional and counterfactual queries, the latter case being possible due to the invertible nature of flows. Finally, throughout a series of experiments on synthetic and real data, the proposed method is shown to outperform current approaches for causal discovery as well as making accurate interventional and counterfactual predictions.

View PDF HTML (experimental) Abstract:Two apparently unrelated fields -- normalizing flows and causality -- have recently received considerable attention in the machine learning community. In this work, we highlight an intrinsic correspondence between a simple family of autoregressive normalizing flows and identifiable causal models. We exploit the fact that autoregressive flow architectures define an ordering over variables, analogous to a causal ordering, to show that they are well-suited to performing a range of causal inference tasks, ranging from causal discovery to making…

saved by

• Eliezer de Souza da Silva

related reading

• Faithful, Interpretable Model Explanations via Causal Abstraction | SAIL Blogai.stanford.edu
• Normalizing flowsbehindml.com
• Bryon Aragam // University of Chicagobryonaragam.com
• 1 Introduction – <span style='font-weight: 700'>Causal Inference</span><br/> <i style='color: #00b7ff'>The Mixtape</i>mixtape.scunning.com
• Normalizing Flows for Probabilistic Modeling and Inferencealphaxiv.org
• Getting In To A Causal Flow - by Ken Acquah - Causal Flowscausalflows.substack.com
• Is causality the missing piece of the AI puzzle?qualcomm.com
• Weakly supervised causal representation learningarxiv.org
• Learning the integral of a diffusion model &#8211; Sander Dielemansander.ai
• Causal Scrubbing: a method for rigorously testing interpretability hypotheses [Redwood Research] — LessWronglesswrong.com
• CausaLab — Can LLM Agents Discover Causal Mechanisms by Experiment?dylanzsz.github.io
• Relating Graph Neural Networks to Structural Causal Modelsarxiv.org