用主动学习提升材料自由能模型的精度与采样效率
Physics- and data-driven Active Learning of neural network representations for free energy functions of materials from statistical mechanics
- 结合物理约束与不确定性采样，优化高维蒙特卡洛空间的采样策略
- 在减少数据点数量的同时，显著降低全局及重点区域的均方误差
- 适用于多种材料体系，尤其适合原子尺度数据驱动的自由能建模
精确的自由能表示对理解材料相变动力学至关重要。本文通过尺度衔接方法，利用基于密度泛函理论（DFT）的蒙特卡洛数据训练神经网络，融入原子级信息构建自由能模型。为优化高维蒙特卡洛空间的采样效率，提出一种主动学习框架，整合空间填充采样、基于不确定性的采样以及物理信息引导的采样。此外，还引入超参数调优、动态采样和新颖性强化等策略。这些方法可灵活组合，在减少所需数据点数量的同时，实现全局或特定区域的均方误差（MSE）最小化。该框架广泛适用于多种材料体系的蒙特卡洛采样。
原文摘要 · Abstract (English)
Accurate free energy representations are crucial for understanding phase dynamics in materials. We employ a scale-bridging approach to incorporate atomistic information into our free energy model by training a neural network on DFT-informed Monte Carlo data. To optimize sampling in the high-dimensional Monte Carlo space, we present an Active Learning framework that integrates space-filling sampling, uncertainty-based sampling, and physics-informed sampling. Additionally, our approach includes methods such as hyperparameter tuning, dynamic sampling, and novelty enforcement. These strategies can be combined to reduce MSE,either globally or in targeted regions of interest,while minimizing the number of required data points. The framework introduced here is broadly applicable to Monte Carlo sampling of a range of materials systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。