Equivariant Masked Position Prediction for Efficient Molecular Representation

TL;DR

This paper introduces EMPP, a self-supervised method for molecular graph representation learning that improves the capture of quantum mechanical features and outperforms existing approaches.

Contribution

The paper proposes a novel equivariant masked position prediction task that leverages physical principles to enhance molecular representations in GNNs.

Findings

EMPP outperforms state-of-the-art self-supervised methods.

It improves the learning of quantum mechanical features.

The approach enhances molecular property prediction accuracy.

Abstract

Graph neural networks (GNNs) have shown considerable promise in computational chemistry. However, the limited availability of molecular data raises concerns regarding GNNs' ability to effectively capture the fundamental principles of physics and chemistry, which constrains their generalization capabilities. To address this challenge, we introduce a novel self-supervised approach termed Equivariant Masked Position Prediction (EMPP), grounded in intramolecular potential and force theory. Unlike conventional attribute masking techniques, EMPP formulates a nuanced position prediction task that is more well-defined and enhances the learning of quantum mechanical features. EMPP also bypasses the approximation of the Gaussian mixture distribution commonly used in denoising methods, allowing for more accurate acquisition of physical properties. Experimental results indicate that EMPP significantly enhances performance of advanced molecular architectures, surpassing state-of-the-art self-supervised approaches. Our code is released in https://github.com/ajy112/EMPP