Current Position: Home > Events&News > Events > Content
DeePKS: a machine learning assisted electronic structure model

Date:2023-03-20 Author: Editor}:材料外事 ClickTimes:

TitleDeePKS: a machine learning assisted electronic structure model

TimeMar. 24, 2023 9:30-10:30

Location: Room 209, School of Materails Science and Engineering

SpeakerLinfeng Zhang,DP Technology;AI for Science Institute

Host Prof. Lijun Zhang

Abstract

We introduce a general machine learning-based framework for building an accurate and widely-applicable energy functional within the framework of generalized Kohn-Sham density functional theory. In particular, we develop a way of training self-consistent models that are capable of taking large datasets from different systems and different kinds of labels. We demonstrate that the functional that results from this training procedure, with the efficiency of cheap density functional models, gives chemically accurate predictions on energy, force, dipole, and electron density for a large class of molecules. In addition, with a seamless integration with ABACUS (https://github.com/deepmodeling/abacus-develop), an open-source electronic structure software package, as well as Bohrium (bohrium.dp.tech), a cloud-native computing platform, DeePKS can be successfully applied to a wide range of problems in condensed phase.

Introduction to the speaker:

Linfeng Zhang is the founder and chief scientist of DP Technology and a researcher at the AI for Science Institute. In 2020, he graduated from the Program in Applied and Computational Mathematics at Princeton University. Linfeng has been focusing on developing machine-learning based physical models for electronic structures, molecular dynamics, and enhanced sampling. He's one of the main developers of DeePMD-kit, a popular deep learning-based open-source software for molecular simulation in physics, chemistry, and materials science. He is a recipient of the 2020 ACM Gordon Bell Prize.

 

 

 

 

 

 

Links