arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.24752physics.chem-phcs.LG
收藏

自适应T-PaiNN:通过经典到量子迁移学习实现数据高效的GNN互原子势开发

Autotuning T-PaiNN: Enabling Data-Efficient GNN Interatomic Potential Development via Classical-to-Quantum Transfer Learning

Materials Science & Engineering, School for the Engineering of Matter, Transport, & Energy, Arizona State University(材料科学与工程系,物质、传输与能源工程学院,亚利桑那州立大学);Chemical Engineering, School for the Engineering of Matter, Transport, & Energy, Arizona State University(化学工程系,物质、传输与能源工程学院,亚利桑那州立大学)

Vivienne Pelletier, Vedant Bhat, Daniel J. Rivera, Steven A. Wilson, Christopher L. Muhich

AI总结:

本文提出T-PaiNN框架,利用经典力场数据提升GNN-MLIP的数据效率,通过预训练和微调实现量子精度,显著降低误差并加速训练。

摘要:

机器学习互原子势(MLIPs),特别是基于图神经网络(GNN)的模型,提供了一条实现接近密度泛函理论(DFT)精度的有希望途径,但其实际应用常受限于所需的大量昂贵量子力学训练数据。本文介绍了一种迁移学习框架,即Transfer-PaiNN(T-PaiNN),通过利用廉价的经典力场数据显著提高了GNN-MLIP的数据效率。该方法包括在大规模经典分子模拟数据集上预训练PaiNN MLIP架构,随后使用相对较小的DFT数据集进行微调(称为自适应)。我们展示了自适应T-PaiNN在气相分子系统(QM9数据集)和液相水中的有效性。在所有情况下,T-PaiNN均显著优于仅使用DFT数据训练的模型,实现数量级的均方误差降低,同时加速训练收敛。例如,使用QM9数据集,在低数据条件下,误差减少高达25倍,而液态水模拟显示能量、力和实验相关性质如密度和扩散的改进预测。这些收益源于模型从大量经典采样中学习势能面的通用特征,随后被精炼到量子精度。总体而言,本文确立了从经典力场迁移学习作为开发高精度、数据高效GNN互原子势的实用且计算高效策略,使MLIPs更广泛应用于复杂化学系统。

英文摘要:

Machine-learned interatomic potentials (MLIPs), particularly graph neural network (GNN)-based models, offer a promising route to achieving near-density functional theory (DFT) accuracy at significantly reduced computational cost. However, their practical deployment is often limited by the large volumes of expensive quantum mechanical training data required. In this work, we introduce a transfer learning framework, Transfer-PaiNN (T-PaiNN), that substantially improves the data efficiency of GNN-MLIPs by leveraging inexpensive classical force field data. The approach consists of pretraining a PaiNN MLIP architecture on large-scale datasets generated from classical molecular simulations, followed by fine-tuning (dubbed autotuning) using a comparatively small DFT dataset. We demonstrate the effectiveness of autotuning T-PaiNN on both gas-phase molecular systems (QM9 dataset) and condensed-phase liquid water. Across all cases, T-PaiNN significantly outperforms models trained solely on DFT data, achieving order-of-magnitude reductions in mean absolute error while accelerating training convergence. For example, using the QM9 data set, error reductions of up to 25 times are observed in low-data regimes, while liquid water simulations show improved predictions of energies, forces, and experimentally relevant properties such as density and diffusion. These gains arise from the model's ability to learn general features of the potential energy surface from extensive classical sampling, which are subsequently refined to quantum accuracy. Overall, this work establishes transfer learning from classical force fields as a practical and computationally efficient strategy for developing high-accuracy, data-efficient GNN interatomic potentials, enabling broader application of MLIPs to complex chemical systems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑