Causal Discovery Workshop

The Causal Discovery Workshop, held in Puebla, Mexico, brings together researchers at the frontier of causal inference, machine learning, and statistical methodology to advance algorithms and frameworks for discovering causal structures from observational and interventional data. Causal discovery — the automated identification of cause-effect relationships from data without controlled experiments — has become a central challenge as data-driven sciences seek explanatory power beyond correlation. The workshop addresses both theoretical foundations, including constraint-based and score-based structure learning algorithms, and applied developments in domains such as genomics, economics, climate science, and healthcare, fostering focused discussions aimed at pushing the boundaries of causal structure learning.

causal discovery causal reasoning graphical models

The Causal Discovery Workshop invites submissions of original research on all aspects of causal structure learning and its applications. Topics of interest include: constraint-based causal discovery algorithms (PC, FCI, RFCI); score-based structure learning (GES, NOTEARS, DAG-GNN); causal discovery with latent confounders; continuous optimization approaches to DAG learning; causal discovery from time-series data; interventional data and experimental design for causal learning; scalable causal discovery for high-dimensional datasets; causal feature selection; evaluation benchmarks for causal discovery algorithms; and applications in biology, medicine, economics, and climate science. Both theoretical and empirical papers are welcomed.