In-Person Poster presentation / poster accept
Learning to Induce Causal Structure
Nan Rosemary Ke · Silvia Chiappa · Jane Wang · Jorg Bornschein · Anirudh Goyal · Melanie Rey · Theophane Weber · Matthew Botvinick · Michael Mozer · Danilo Jimenez Rezende
MH1-2-3-4 #62
Keywords: [ Deep Learning and representational learning ] [ causality ] [ deep learning ]
The fundamental challenge 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.