<ul><li>Causal discovery, the task of inferring causal structure from data, has the potential to uncover mechanistic insights from biological experiments.</li><li>To address challenges in causal discovery with larger sets of variables and limited data, a foundation model-inspired approach is proposed.</li><li>The approach involves training a supervised model on large-scale, synthetic data to predict causal graphs from summary statistics.</li><li>Experiments show that the model generalizes well, runs on graphs with hundreds of variables in seconds, and is adaptable to different data assumptions.</li></ul>

Sample, estimate, aggregate: A recipe for causal discovery foundation models

Discover more