Graph Transformer Networks for Accurate Band Structure Prediction: An End-to-End Approach

2024-11-25

arXiv.org

AI Summary

Accurately predicting electronic band structures directly from crystal structures remains a challenging end-to-end task in computational materials science. Method: This work introduces the first deep learning model that directly maps atomic graph representations of crystals to continuous *k*-path band sequences. It innovatively integrates a graph Transformer with continuous *k*-space modeling, treating band structures as ordered sequences for joint optimization—bypassing computationally expensive first-principles calculations or indirect Hamiltonian construction. Contribution/Results: Given input crystal graphs (atomic species, 3D coordinates, and bonding topology), the model simultaneously predicts high-fidelity band dispersion curves and key electronic properties—including bandgap, band center, and bandwidth. Evaluated on a large-scale, chemically diverse crystal dataset, it achieves a mean absolute error (MAE) of <0.05 eV for band energies and <0.03 eV for bandgaps, substantially outperforming existing machine learning approaches. This establishes a new paradigm for rapid, accurate electronic structure screening of novel materials.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Predicting electronic band structures from crystal structures is crucial for understanding structure-property correlations in materials science. First-principles approaches are accurate but computationally intensive. Recent years, machine learning (ML) has been extensively applied to this field, while existing ML models predominantly focus on band gap predictions or indirect band structure estimation via solving predicted Hamiltonians. An end-to-end model to predict band structure accurately and efficiently is still lacking. Here, we introduce a graph Transformer-based end-to-end approach that directly predicts band structures from crystal structures with high accuracy. Our method leverages the continuity of the k-path and treat continuous bands as a sequence. We demonstrate that our model not only provides accurate band structure predictions but also can derive other properties (such as band gap, band center, and band dispersion) with high accuracy. We verify the model performance on large and diverse datasets.

Problem

• Predicting electronic band structures from crystal structures accurately and efficiently
• Overcoming computational intensity of first-principles band structure calculations
• Developing end-to-end machine learning model for direct band structure prediction

Innovation

• Graph Transformer-based end-to-end band structure prediction
• Treats continuous bands as sequence on k-path
• Derives multiple electronic properties from crystal structures