Prime Implicant Explanations for Reaction Feasibility Prediction

2025-10-10

AI Summary

To address the limited interpretability of machine learning models in chemical reaction feasibility prediction, this work introduces prime implicant explanations—novel to this domain—and defines a minimal sufficient reason tailored to molecular graph structures. Our method integrates symbolic computation with subgraph analysis, modeling atom- and bond-level feature dependencies via chemical reaction rules to automatically extract the most concise critical substructures determining prediction outcomes. Experiments demonstrate stable identification of core structural features governing reaction feasibility; the explanations are chemically meaningful and conservatively faithful. Validated on small-scale tasks, the approach achieves both high interpretability and competitive predictive accuracy. Key contributions are: (1) the first prime implicant explanation framework for reaction feasibility prediction; (2) a formally grounded definition of minimal sufficient reason adapted to molecular graphs; and (3) a principled trade-off between model transparency and predictive performance.

Technology Category

• Reasoning under Uncertainty: Graphical Models
• Machine Learning: Probabilistic Circuits and Graphical Models
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Machine learning models that predict the feasibility of chemical reactions have become central to automated synthesis planning. Despite their predictive success, these models often lack transparency and interpretability. We introduce a novel formulation of prime implicant explanations--also known as minimally sufficient reasons--tailored to this domain, and propose an algorithm for computing such explanations in small-scale reaction prediction tasks. Preliminary experiments demonstrate that our notion of prime implicant explanations conservatively captures the ground truth explanations. That is, such explanations often contain redundant bonds and atoms but consistently capture the molecular attributes that are essential for predicting reaction feasibility.

Problem

• Predicting chemical reaction feasibility lacks transparency and interpretability in models
• Introducing prime implicant explanations tailored for reaction prediction tasks
• Capturing essential molecular attributes while identifying redundant bonds and atoms

Innovation

• Prime implicant explanations tailored for chemical reactions
• Algorithm computes minimal sufficient reasons for predictions
• Conservatively captures essential molecular attributes for feasibility