arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2006.05475physics.comp-phcond-mat.mtrl-scistat.ML
收藏

简单有效的算法,用于训练机器学习潜力,以迫使数据

Simple and efficient algorithms for training machine learning potentials to force data

Justin S. Smith, Nicholas Lubbers, Aidan P. Thompson, Kipton Barros

摘要:

摘要机器学习模型,从从头算量子模拟的数据训练,产生分子动力学潜力与前所未有的准确性。一个限制因素是可用训练数据的数量,这可能是昂贵的。一个量子模拟通常提供所有的原子力,除了系统的总能量。这些力提供的信息比单独的能量多得多。可能出现的是,针对如此大量的力数据训练模型将引入显著的计算成本。实际上,训练所有可用的力数据应该只比单独训练能量贵几倍。在这里,我们提出了一种用于有效力训练的新算法,并通过训练来自有机化学和散装铝的真实数据集的力来衡量其准确性。

英文摘要:

Abstract Machine learning models, trained on data from ab initio quantum simulations, are yielding molecular dynamics potentials with unprecedented accuracy. One limiting factor is the quantity of available training data, which can be expensive to obtain. A quantum simulation often provides all atomic forces, in addition to the total energy of the system. These forces provide much more information than the energy alone. It may appear that training a model to this large quantity of force data would introduce significant computational costs. Actually, training to all available force data should only be a few times more expensive than training to energies alone. Here, we present a new algorithm for efficient force training, and benchmark its accuracy by training to forces from real-world datasets for organic chemistry and bulk aluminum.

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。

↑