Meta-Learning Linear Models for Molecular Property Prediction

2025-09-16

AI Summary

Chemical property prediction faces the dual challenge of limited small-sample data and the difficulty of balancing predictive performance with model interpretability. To address this, we propose a meta-learning-driven linear interpretable method: leveraging multi-task meta-learning to identify task-shared parameters and construct a universal functional manifold for initializing new tasks—enabling knowledge transfer without sharing raw data. By integrating ridge regression with multi-task learning, the model preserves strict linearity, ensuring transparency and post-hoc explainability. Evaluated on multiple molecular property prediction benchmarks, our approach achieves 1.1×–25× higher predictive accuracy than conventional linear models, markedly improving reliability and interpretability in low-data regimes. This work establishes a novel paradigm for eXplainable AI (XAI)-enabled chemical modeling, bridging the gap between data efficiency, generalization, and scientific interpretability.

Technology Category

• Machine Learning: Transparent, Interpretable, Explainable ML
• Computer Vision: Interpretability, Explainability, and Transparency
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Chemists in search of structure-property relationships face great challenges due to limited high quality, concordant datasets. Machine learning (ML) has significantly advanced predictive capabilities in chemical sciences, but these modern data-driven approaches have increased the demand for data. In response to the growing demand for explainable AI (XAI) and to bridge the gap between predictive accuracy and human comprehensibility, we introduce LAMeL - a Linear Algorithm for Meta-Learning that preserves interpretability while improving the prediction accuracy across multiple properties. While most approaches treat each chemical prediction task in isolation, LAMeL leverages a meta-learning framework to identify shared model parameters across related tasks, even if those tasks do not share data, allowing it to learn a common functional manifold that serves as a more informed starting point for new unseen tasks. Our method delivers performance improvements ranging from 1.1- to 25-fold over standard ridge regression, depending on the domain of the dataset. While the degree of performance enhancement varies across tasks, LAMeL consistently outperforms or matches traditional linear methods, making it a reliable tool for chemical property prediction where both accuracy and interpretability are critical.

Problem

• Addressing limited high-quality datasets for molecular property prediction
• Bridging the gap between predictive accuracy and interpretability in chemistry
• Leveraging meta-learning to improve performance across multiple chemical properties

Innovation

• Meta-learning linear models for interpretability
• Shared parameters across tasks without data sharing
• Consistently outperforms traditional linear methods