Active Causal Machine Learning for Molecular Property Prediction

Zachary R Fox, Ayana Ghosh

Submission Track: Findings

Submission Category: AI-Guided Design + Automated Chemical Synthesis

Keywords: Active learning, causal ML, molecular property prediction

TL;DR: We developed an active learning algorithm to reconstruct global causal molecular property relationships from a minimal dataset.

Abstract: Predicting properties from molecular structures is paramount to design tasks in medicine, materials science, and environmental management. However, design rules derived from the structure-property relationships using correlative data-driven methods fail to elucidate underlying causal mechanisms controlling chemical phenomena. This preliminary work proposes a workflow to actively learn robust cause-effect relations between structural features and molecular property for a broad chemical space utilizing smaller subsets, entailing partial information.

Digital Discovery Special Issue: Yes

Submission Number: 48