|Name||Center for Theoretical Physics Colloquium|
|Title||Phase Detection with Neural Networks: Interpreting the Black Box|
Neural networks (NNs) normally do not allow any insight into the reasoning behind their predictions. We demonstrate how influence functions can unravel the black box of NN when trainedto predict the phases of the one-dimensional extended spinless Fermi-Hubbard model at half-filling.Results are the first indication that the NN correctly learns an order parameter describing thetransition. Moreover, we demonstrate that influence functions not only allow to check that the network trained to recognize known quantum phases can predict new unknown ones but even disclosesinformation about the type of phase transition.
|Time||Wednesday, 22 April 2020, at 12:30 CEST The seminar was held!|
Anna Dawid (Faculty of Physics, University of Warsaw, Warsaw, Poland & ICFO-Institut de Ciencies Fotoniques, Castelldefels (Barcelona), Spain)
|Organisers||Maciej Bilicki; Adam Sawicki; Krzysztof Pawłowski; Julius Serbenta;|