Transferable FB-GNN-MBE Framework for Potential Energy Surfaces: Data-Adaptive Transfer Learning in Deep Learned Many-Body Expansion Theory
2026-04-10 • Machine Learning
Machine Learning
AI summaryⓘ
The authors created a new method called FB-GNN-MBE that predicts how molecules interact by combining a fragment-based graph neural network with a many-body expansion approach. They break big molecules into smaller pieces, calculate energies for single fragments using quantum methods, and use the neural network to estimate interactions between fragments. Their approach accurately predicts energies in water and phenol systems and can handle large molecules efficiently. They also used a teacher-student learning setup to transfer knowledge from a complex model to a simpler one, making predictions faster with less data. Overall, their method improves accuracy and practicality for simulating large molecular systems.
Fragment-based methodGraph neural network (GNN)Many-body expansion (MBE)Potential energy surface (PES)Quantum mechanics (QM)Transfer learningMolecular simulationTwo-body interactionThree-body interactionDissociation curve
Authors
Siqi Chen, Zhiqiang Wang, Yili Shen, Xianqi Deng, Xi Cheng, Cheng-Wei Ju, Jun Yi, Guo Ling, Dieaa Alhmoud, Hui Guan, Zhou Lin
Abstract
Mechanistic understanding and rational design of complex chemical systems depend on fast and accurate predictions of electronic structures beyond individual building blocks. However, if the system exceeds hundreds of atoms, first-principles quantum mechanical (QM) modeling becomes impractical. In this study, we developed FB-GNN-MBE by integrating a fragment-based graph neural network (FB-GNN) into the many-body expansion (MBE) theory and demonstrated its capacity to reproduce first-principles potential energy surfaces (PES) for hierarchically structured systems with manageable accuracy, complexity, and interpretability. Specifically, we divided the entire system into basic building blocks (fragments), evaluated their one-fragment energies using a QM model, and addressed many-fragment interactions using the structure-property relationships trained by FB-GNNs. Our investigation shows that FB-GNN-MBE achieves chemical accuracy in predicting two-body (2B) and three-body (3B) energies across water, phenol, and mixture benchmarks, as well as the one-dimensional dissociation curves of water and phenol dimers. To transfer the success of FB-GNN-MBE across various systems with minimal computational costs and data demands, we developed and validated a teacher-student learning protocol. A heavy-weight FB-GNN trained on a mixed-density water cluster ensemble (teacher) distills its learned knowledge and passes it to a light-weight GNN (student), which is later fine-tuned on a uniform-density (H2O)21 cluster ensemble. This transfer learning strategy resulted in efficient and accurate prediction of 2B and 3B energies for variously sized water clusters without retraining. Our transferable FB-GNN-MBE framework outperformed conventional non-FB-GNN-based models and showed high practicality for large-scale molecular simulations.