科研交流
您的位置>> 首 页 > 科研交流 > 学术交流 > 正文
学术交流

大数据科学训练营之《深度学习在原子建模中的应用》

发表于: 2018-04-16 14:08  点击:

讲座题目:大数据科学训练营之《深度学习在原子建模中的应用

讲座时间: 4月21日(周六)13:30-15:00

讲座地点: 吉林大学中心校区李四光教学楼210

主讲人: 王涵

北京大学计算数学博士

北京应用物理与计算数学研究所副研究员

德国马克思-普朗克研究所博士后研究员

适应对象:各专业本科生、研究生、教师

具体内容:In this talk, two classical problems in molecular dynamics, modeling the atomic interaction and the computing the free energy, are studied with the deep learning technique. The Deep Potential Molecular Dynamics (DeePMD) method is based on a many-body potential generated by a carefully crafted deep neural network trained with ab initio data. The neural network model preserves all the natural symmetries in the atomic energy. It is "first principle-based" in the sense that there are no ad hoc components aside from the network model. We show that the proposed scheme provides an efficient and accurate protocol in a variety of systems, including bulk materials and molecules. The reinforced dynamics enhances the sampling of the configurational space and computes the free energy by applying a biasing potential, which is adaptively trained with data collected judiciously from the exploration. Parameterization using neural networks makes it feasible to handle cases with a large set of collective variables. This has the potential advantage that selecting precisely the right set of collective variables has now become less critical for capturing the structural transformations of the system. The method is illustrated by studying the full-atom, explicit solvent models of alanine dipeptide and tripeptide, as well as the system of a polyalanine-10 molecule with 20 collective variables.

主办单位:数学学院、国家天元数学东北中心

系列报告信息欢迎大家关注“大数据科学训练营”微信公众号

  • 附件【图片1.png】已下载