Join the Learning on Graphs and Geometry Reading Group: https://hannesstark.com/logagreadin...
Paper "Relating Graph Neural Networks to Structural Causal Models": https://arxiv.org/pdf/2109.04173.pdf
Abstract: Causality can be described in terms of a structural causal model (SCM) that carries information on the variables of interest and their mechanistic relations. For most processes of interest the underlying SCM will only be partially observable, thus causal inference tries to leverage any exposed information. Graph neural networks (GNN) as universal approximators on structured input pose a viable candidate for causal learning, suggesting a tighter integration with SCM. To this effect we present a theoretical analysis from first principles that establishes a novel connection between GNN and SCM while providing an extended view on general neuralcausal models. We then establish a new model class for GNNbased causal inference that is necessary and sufficient for causal effect identification. Our empirical illustration on simulations and standard benchmarks validate our theoretical proofs.
Authors: Matej Zečević, Devendra Singh Dhami, Petar Veličković, Kristian Kersting
Links:
Twitter Hannes: / hannesstaerk
Twitter Dominique: / dom_beaini
Twitter Valence Discovery: / valence_ai
Reading Group Slack: https://join.slack.com/t/logag/shared...
~
00:00 Introduction
01:13 Towards NeuroCausality intro
25:28 GNNBased Causal Inference
48:35 Identifiability & Estimation
59:55 Causality for Machine Learning
1:04:20 Q&A