Machine learning in condensed matter physics

TKM Institutsseminar

Vortragender:

Dr. Elmer V. H. Doggen

Datum:

23.11.2017 12:30

Ort:

Room 10.01, 10th Floor, Bldg. 30.23, KIT Campus South

Zugehörigkeit:

Karlsruhe Institute of Technology

Gastgeber:

Prof. Dr. Alexander Mirlin

Abstract

Machine learning is emerging as a promising new tool in the computational toolbox. I will briefly explain the key concept of a restricted Boltzmann machine and discuss the application of machine learning to I) solving quantum many-body problems and II) distinguishing phases from raw data, as presented in two recent papers:

I) G. Carleo and M. Troyer, Solving the quantum many-body problem with artificial neural networks, Science 355, 602 (2017)
II) F. Schindler, N. Regnault, T. Neupert, Probing many-body localization with neural networks, Phys. Rev. B 95, 245134 (2017)