PDDFormer: Pairwise Distance Distribution Graph Transformer for Crystal Material Property Prediction
This work solves the problem of accurate crystal property prediction for materials science researchers, representing an incremental improvement over existing pairwise distance distribution methods.
The paper tackles the challenge of predicting crystal material properties by addressing discontinuities in traditional crystal graphs caused by atomic vibrations, introducing atom-Weighted Pairwise Distance Distribution (WPDD) and Unit cell Pairwise Distance Distribution (UPDD) to construct multi-edge crystal graphs. The result is WPDDFormer achieving state-of-the-art predictive accuracy on benchmark datasets like the Materials Project and JARVIS-DFT.
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.