LoSAM: Local Search in Additive Noise Models with Mixed Mechanisms and General Noise for Global Causal Discovery

Sujai Hiremath, PROMIT GHOSAL, Kyra Gan

Keywords: causal discovery, additive noise model, functional causal model, mixed mechanism

TL;DR: We develop a novel functional causal model based method that handles additive noise models with mixed mechanisms and general noise, loosening the assumptions made by prior work and achieving sota performance in numerical experiments.

Abstract:

Inferring causal relationships from observational data is crucial when experiments are costly or infeasible. Additive noise models (ANMs) enable unique directed acyclic graph (DAG) identification, but existing sample-efficient ANM methods often rely on restrictive assumptions on the data generating process, limiting their applicability to real-world settings.
We propose local search in additive noise models, LoSAM, a topological ordering method for learning a unique DAG in ANMs with mixed causal mechanisms and general noise distributions.
We introduce new causal substructures and criteria for identifying roots and leaves, enabling efficient top-down learning. We prove asymptotic consistency and polynomial runtime, ensuring scalability and sample efficiency. We test LoSAM on synthetic and real-world data, demonstrating state-of-the-art performance across all mixed mechanism settings.

Code Link: https://github.com/Sujai1/local-search-discovery

Readers: auai.org/UAI/2025/Conference, auai.org/UAI/2025/Conference/Area_Chairs, auai.org/UAI/2025/Conference/Reviewers, auai.org/UAI/2025/Conference/Submission836/Authors, auai.org/UAI/2025/Conference/Submission836/Reproducibility_Reviewers

Submission Number: 836