Learning Neural Causal Models from Unknown Interventions
This addresses a key limitation in causal inference for researchers by enabling structure learning from richer interventional data without needing prior knowledge of interventions, though it is incremental as it builds on existing optimization methods.
The paper tackles the problem of learning causal structures from both observational and interventional data, even when the intervened variable is unknown, by proposing a neural network-based continuous optimization framework, achieving strong benchmark results on synthetic and standard graphs.
Promising results have driven a recent surge of interest in continuous optimization methods for Bayesian network structure learning from observational data. However, there are theoretical limitations on the identifiability of underlying structures obtained from observational data alone. Interventional data provides much richer information about the underlying data-generating process. However, the extension and application of methods designed for observational data to include interventions is not straightforward and remains an open problem. In this paper we provide a general framework based on continuous optimization and neural networks to create models for the combination of observational and interventional data. The proposed method is even applicable in the challenging and realistic case that the identity of the intervened upon variable is unknown. We examine the proposed method in the setting of graph recovery both de novo and from a partially-known edge set. We establish strong benchmark results on several structure learning tasks, including structure recovery of both synthetic graphs as well as standard graphs from the Bayesian Network Repository.