Boosting Crystalline Property Prediction through Dynamical Feature Updating and Wavelet-Denoised Features: A New Deep Learning Model
- 19 February 2026
- journal article
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry Letters
- Vol. 17 (9) , 2552-2558
Abstract
Graph neural networks (GNNs) inherently excel at representing atomistic material, therefore enabling predictions of properties with high accuracy with embedded descriptors. Approaching to accurate prediction of a variety of physical properties of diverse materials, particularly those with multicomponents, remains challenging due to the complexity of many-body atomic interactions and the reliance on handcrafted symmetry descriptors. Here, we introduce the Wavelet Atomic Neighborhood Network (WANN), a novel framework that implicitly captures many-body interactions through an iterative subembedding module for updating atomic features, thereby eliminating the integration of the predefined feature engineering. A wavelet-based regression component is built up through multiscale feature analysis. WANN demonstrates superior accuracy, achieving small mean absolute error (MAE) in predicting a variety of physical and chemical quantities, including thermodynamical, dielectric, piezoelectric, magnetic, and thermoelectric properties. For instance, it could significantly reduce MAE maximum by 61.64% in comparison to Matformer for shear moduli and by 90.84% in comparison to ALIGNN for the exfoliation energy. This scalable and preprocessing-light deep-learning framework provides an efficient toolkit for facilitating high-throughput screening of emergent materials, like high-entropy alloys and ceramics.Keywords
Funding Information
- Southeast University
This publication has 29 references indexed in Scilit:
- X2 ‐ PEC : A Neural Network Model Based on Atomic Pair Energy CorrectionsJournal of Computational Chemistry, 2025
- X2-GNN: A Physical Message Passing Neural Network with Natural Generalization Ability to Large and Complex MoleculesThe Journal of Physical Chemistry Letters, 2024
- Constructing Accurate and Efficient General-Purpose Atomistic Machine Learning Model with Transferable Accuracy for Quantum ChemistryJournal of Chemical Theory and Computation, 2024
- MEGnet: Automatic ICA-based artifact removal for MEG using spatiotemporal convolutional neural networksNeuroImage, 2021
- Gaussian Process Regression for Materials and MoleculesChemical Reviews, 2021
- Random-Phase Approximation in Many-Body Noncovalent Systems: Methane in a Dodecahedral Water CageJournal of Chemical Theory and Computation, 2021
- The joint automated repository for various integrated simulations (JARVIS) for data-driven materials designnpj Computational Materials, 2020
- Performance and Cost Assessment of Machine Learning Interatomic PotentialsThe Journal of Physical Chemistry A, 2020
- Recent advances and applications of machine learning in solid-state materials sciencenpj Computational Materials, 2019
- Python Materials Genomics (pymatgen): A robust, open-source python library for materials analysisComputational Materials Science, 2013