Learning to induce causal structure

Rosemary Ke
Silvia Chiappa
Jane Wang
Anirudh Goyal
Jörg Bornschein
Melanie Rey
Theophane Weber
Matthew Botvinick
Mike Mozer
Danilo Jimenez Rezende
International Conference on Learning Representations(2023)


One of the fundamental challenges in causal induction is to infer the underlying graph structure given observational and/or interventional data. Most existing causal induction algorithms operate by generating candidate graphs and evaluating them using either score-based methods (including continuous optimization) or independence tests. In our work, we instead treat the inference process as a black box and design a neural network architecture that learns the mapping from both observational and interventional data to graph structures via supervised training on synthetic graphs. The learned model generalizes to new synthetic graphs, is robust to train-test distribution shifts, and achieves state-of-the-art performance on naturalistic graphs for low sample complexity.

Research Areas