Learning to Induce Causal Structure
This work addresses the causal induction problem for researchers and practitioners in fields like machine learning and statistics, offering a novel approach that is robust and generalizable, though it builds on existing methods.
The paper tackles the challenge of inferring causal graph structures from observational and interventional data by proposing a neural network that learns this mapping through supervised training on synthetic graphs, achieving state-of-the-art performance on naturalistic graphs with low sample complexity.
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.