Speaker
Evert van Nieuwenburg
(California Institute of Technology)
Description
This work shows that it is possible to detect phase transitions in condensed matter systems using a novel approach based on machine learning. A neural network is trained on purposefully (mis)labeled data, after which the transition point can be identified from the network's performance. This technique is capable of identifying thermodynamic and topological transitions as well as other non-trivial transitions such as in many-body localization.
