Integrates Kolmogorov-Arnold Networks (KANs) into causal generative modeling to produce human-readable symbolic structural equations.
March 23, 2026
Original Paper
Kolmogorov-Arnold causal generative models
arXiv · 2603.20184
The Takeaway
Moves beyond black-box causal models by using KANs to explicitly visualize parent-child relationships and extract symbolic approximations. This bridges the gap between high-expressivity deep learning and the interpretability required for high-stakes tabular decision-making.
From the abstract
Causal generative models provide a principled framework for answering observational, interventional, and counterfactual queries from observational data. However, many deep causal models rely on highly expressive architectures with opaque mechanisms, limiting auditability in high-stakes domains. We propose KaCGM, a causal generative model for mixed-type tabular data where each structural equation is parameterized by a Kolmogorov--Arnold Network (KAN). This decomposition enables direct inspection