PRISM: Periodic Representation with multIscale and Similarity graph Modelling for enhanced crystal structure property prediction
This work solves the problem of accurate crystal property prediction for materials science researchers, representing an incremental advancement by integrating specialized modules into a graph neural network framework.
The paper tackled the problem of predicting crystal structure properties by addressing the limitations of existing graph-based methods in handling periodic boundary conditions and multiscale interactions, resulting in improved state-of-the-art predictive accuracy as demonstrated in benchmarks.
Crystal structures are characterised by repeating atomic patterns within unit cells across three-dimensional space, posing unique challenges for graph-based representation learning. Current methods often overlook essential periodic boundary conditions and multiscale interactions inherent to crystalline structures. In this paper, we introduce PRISM, a graph neural network framework that explicitly integrates multiscale representations and periodic feature encoding by employing a set of expert modules, each specialised in encoding distinct structural and chemical aspects of periodic systems. Extensive experiments across crystal structure-based benchmarks demonstrate that PRISM improves state-of-the-art predictive accuracy, significantly enhancing crystal property prediction.