New DAGMA method improves structure learning with varying disturbances

Researchers have developed a new method for learning Directed Acyclic Graph (DAG) structures from data, specifically addressing challenges posed by disturbances that vary in scale and tail behavior across nodes. The proposed node-varying Student-t DAGMA model jointly estimates the graph structure, individual node disturbance scales, and a degrees-of-freedom parameter. This approach utilizes a Gaussian scale-mixture representation for an iteratively reweighted formulation that effectively down-weights large residuals, leading to improved structural recovery in synthetic experiments with heterogeneous heavy-tailed disturbances. Real-world finance data experiments showed the method achieved lower test negative log-likelihood compared to existing baselines. AI

IMPACT Enhances the robustness of causal discovery methods in the presence of complex, non-Gaussian data distributions.

RANK_REASON Academic paper on a novel methodology for DAG structure learning. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• Dagmar
• DagsHub
• Gaussian function
• Gotit.pub
• Hugging Face
• IArxiv
• ScienceCast
• Student's t-test

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →