Prequential MDL for Causal Structure Learning with Neural Networks

Abstract

Learning the structure of Bayesian networks and causal relationships from observations is a common goal in several areas of science and technology. We show that the prequential minimum description length principle (MDL) can be used to derive a practical scoring function for Bayesian networks when flexible and overparametrized neural networks are used to model the conditional probability distributions between observed variables. MDL represents an embodiment of Occam's Razor and we obtain plausible and parsimonious graph structures without relying on sparsity inducing priors or other regularizers which must be tuned. Empirically we demonstrate competitive results on synthetic and real-world data. The score often recovers the correct structure even in the presence of strongly nonlinear relationships between variables; a scenario were prior approaches struggle and usually fail. Furthermore we discuss how the the prequential score relates to recent work that infers causal structure from the speed of adaptation when the observations come from a source undergoing distributional shift.

References in corpus (9)

• A tutorial introduction to the minimum description length principle
• DAGs with NO TEARS: Continuous Optimization for Structure Learning
• Bayesian network learning with cutting planes
• DAG-GNN: DAG Structure Learning with Graph Neural Networks
• Learning Neural Causal Models from Unknown Interventions
• The Description Length of Deep Learning Models
• Asymptotics of Discrete MDL for Online Prediction
• Quotient Normalized Maximum Likelihood Criterion for Learning Bayesian Network Structures
• Small Data, Big Decisions: Model Selection in the Small-Data Regime