报告题目：Modeling the interatomic potential by deep learning

报告人：王涵 副研究员,北京应用物理与计算数学研究所

报告时间：2021年1月8日（周五）上午10:00

报告地点：腾讯会议ID: 582 466 169

会议直播网址：https://meeting.tencent.com/l/ngip3vRjOVGJ

会议直播网址二维码：

报告摘要：

In silico design of materials requires an accurate description of the interatomic potential energy surface (PES). However, in the context of molecular simulation, one usually faces the dilemma that the first principle PESs are accurate but computationally expensive, while the empirical PESs (force fields) are efficient but of limited accuracy. We discuss the solution in two aspects: PES construction and data generation. In terms of PES construction, we introduce the Deep Potential (DP) method, which faithfully represents the first principle PES by a symmetry-preserving deep neural network. In terms of data generation, we present a new concurrent learning scheme named Deep Potential Generator (DP-GEN). This approach automatically generates the most compact training dataset that enables the training of DP with uniform accuracy. By contrast to the empirical PESs, the DP-GEN opens the opportunity of continuously improving the quality of DP by exploring the chemical and configurational space of the system. We briefly introduce the open-source implementations of DP and DP-GEN and then debut the DP library that provides the platform for openly sharing DP models and data in the community. In the last part of the talk, we briefly introduce the optimization of DeePMD-kit on Summit supercomputer, which wins the 2020 ACM Gordon Bell prize for its unprecedented power of simulating 100M atoms with the first-principle accuracy in one day.

报告人简介：

王涵于2011年获得北京大学理学博士学位，之后前往柏林自由大学进行博士后研究。2014年加入北京应用物理与计算数学研究所，现任副研究员，博士生导师。王涵的主要研究兴趣为分子模拟中的多尺度建模与计算方法。王涵2019年入选北京市青年人才托举工程，并获得中国数学会计算数学分会第五届青年创新奖。

邀请人：袁声军教授