论文标题
有效学习精确的替代物以模拟复杂系统
Efficient Learning of Accurate Surrogates for Simulations of Complex Systems
论文作者
论文摘要
机器学习方法越来越多地用于为复杂的物理模型构建计算廉价的替代物。当数据嘈杂，稀疏或时间依赖性时，这些替代物的预测能力会受到影响。由于我们有兴趣找到一个为任何潜在的未来模型评估提供有效预测的替代物，因此我们引入了一种在线学习方法，该方法由优化器驱动的采样授权。该方法比当前方法具有两个优点。首先，它确保训练数据中包括模型响应表面上的所有转弯点。其次，在进行任何新的模型评估之后，如果“得分”降至有效性阈值以下，则对替代物进行测试并“重新训练”（更新）。基准功能的测试表明，以优化器指导的采样通常优于传统抽样方法，即使评分度量指标有利于总体准确性，就本地极值周围的准确性而言。我们将我们的方法应用于核物质的模拟，以证明可以使用一些模型评估可以从昂贵的计算中可靠地自动生成状态核方程的替代物。
Machine learning methods are increasingly used to build computationally inexpensive surrogates for complex physical models. The predictive capability of these surrogates suffers when data are noisy, sparse, or time-dependent. As we are interested in finding a surrogate that provides valid predictions of any potential future model evaluations, we introduce an online learning method empowered by optimizer-driven sampling. The method has two advantages over current approaches. First, it ensures that all turning points on the model response surface are included in the training data. Second, after any new model evaluations, surrogates are tested and "retrained" (updated) if the "score" drops below a validity threshold. Tests on benchmark functions reveal that optimizer-directed sampling generally outperforms traditional sampling methods in terms of accuracy around local extrema, even when the scoring metric favors overall accuracy. We apply our method to simulations of nuclear matter to demonstrate that highly accurate surrogates for the nuclear equation of state can be reliably auto-generated from expensive calculations using a few model evaluations.
