Discover Quantum Phase Transition by Deep Learning
Chi-Ting Ho1*, Daw-Wei Wang1,2
1Department of physics, National Tsing Hua University, Hsinchu, Taiwan
2Physics Division, National Center for Theoretical Sciences, Hsinchu, Taiwan
* Presenter:Chi-Ting Ho, email:aeio6646@yahoo.com.tw
It has been believed that the phase boundary between two different quantum
states can be determined by a machine learning model which is trained by the
experimental data in the deep regimes of these two phases. However, this
approach may not be reliable because it is still based on a theoretical work
that a phase transition must exist somewhere in-between. In this work, we
develop a new training method, which could be applied in a much wider
situation, so that our Deep Learning model could determine if there could be
a quantum/topological phase transition from experimental data directly even
without assuming its existence. Our method could be in principle also
applied to discover new many-body states even without any a priori
theoretical works.


Keywords: Machine learning, Deep learning, Phase transition