Careers | Phone Book | A - Z Index

Yu-Hang Tang

Yu Hang Tang head
Dr. Yu-Hang (Maxin) Tang
Research Scientist
MS #50F-1645
1 Cyclotron Rd
Berkeley, CA 94720 US

Postdoctoral fellow opening --- click here.

Yu-Hang Tang is a Research Scientist with the Computational Chemistry, Materials and Climate Group in LBL's Computational Research Division.  He joined the lab in Fall 2017 as an Alvarez Postdoctoral Fellow. With formal training in applied mathematics and polymer science, he has marched along the interface between mathematics, computer science, and natural science, seeking interesting topics concerning methodology development for data-driven modeling of chemistry and material systems.

His other long-time research interests include multiscale simulation, concurrently coupling, accelerated/heterogeneous computing, and data visualization. He's also the primary author/contributor of several open-source scientific packages such as the Multiscale Universal Interface (MUI), OpenRBC, uDeviceX, and the USER-MESO package for LAMMPS, and a finalist of the 2015 Gordon Bell Award.

Prior to joining LBNL, he obtained a Ph.D. degree in applied mathematics from Brown University under the supervision of Prof. George Karniadakis. Earlier, he was initiated into computer modeling by Prof. Jun Ling while completing his bachelor's degree in polymer materials and engineering at Zhejiang University in China.

He has published more than 20 research papers and has given tens of presentations and lectures at international conferences.

Software project:

GraphDot: GPU-accelerated library for graph similarity computation. Repository Documentation Status

Recent presentations:

A High-Throughput Solver for Marginalized Graph Kernels on GPU. IPDPS 2020. slides

Kernel methods for active learning in the chemical space. 2019 DSI Workshop. DOI





Journal Articles

Yu-Hang Tang, Wibe A. de Jong, "Prediction of atomization energy using graph kernel and active learning", The Journal of Chemical Physics, January 25, 2019, 150:044107, doi: 10.1063/1.5078640

Yu-Hang Tang, Dongkun Zhang, George Em Karniadakis, "An atomistic fingerprint algorithm for learning ab initio molecular force fields", Journal of Chemical Physics, 2018, 148,

Yu-Hang Tang, Shuhei Kudo, Xin Bian, Zhen Li, George Em Karniadakis, "Multiscale Universal Interface: A concurrent framework for coupling heterogeneous solvers", Journal of Computational Physics, September 15, 2015, 297:13-31, doi: 10.1016/