PDDFormer: Pairwise Distance Distribution Graph Transformer for Crystal Material Property Prediction

TL;DR

This paper introduces WPDDFormer, a novel graph transformer that uses pairwise distance distributions to improve crystal property prediction, addressing the issues of discontinuity and high computational costs in traditional methods.

Contribution

The paper proposes atom-Weighted and Unit cell PDDs integrated into a graph transformer, enhancing stability and efficiency in crystal property prediction.

Findings

Achieves state-of-the-art accuracy on Materials Project and JARVIS-DFT datasets.

Demonstrates robustness of crystal graphs under atomic perturbations.

Reduces computational costs compared to traditional PDD methods.

Abstract

Crystal structures can be simplified as a periodic point set that repeats across three-dimensional space along an underlying lattice. Traditionally, crystal representation methods characterize the structure using descriptors such as lattice parameters, symmetry, and space groups. However, in reality, atoms in materials always vibrate above absolute zero, causing their positions to fluctuate continuously. This dynamic behavior disrupts the fundamental periodicity of the lattice, making crystal graphs based on static lattice parameters and conventional descriptors discontinuous under slight perturbations. Chemists proposed the pairwise distance distribution (PDD) method to address this problem. However, the completeness of PDD requires defining a large number of neighboring atoms, leading to high computational costs. Additionally, PDD does not account for atomic information, making it challenging to apply it directly to crystal material property prediction tasks. To tackle these challenges, we introduce the atom-Weighted Pairwise Distance Distribution (WPDD) and Unit cell Pairwise Distance Distribution (UPDD) and apply them to the construction of multi-edge crystal graphs. We demonstrate the continuity and general completeness of crystal graphs under slight atomic position perturbations. Moreover, by modeling PDD as global information and integrating it into matrix-based message passing, we significantly reduce computational costs. Comprehensive evaluation results show that WPDDFormer achieves state-of-the-art predictive accuracy across tasks on benchmark datasets such as the Materials Project and JARVIS-DFT.