Importance Sampling of Complex Probability Distributions

Lukas Kades
Ruprecht-Karls-Universität Heidelberg
Institute for Theoretical Physics

After a short introduction to Quantum Chromo Dynamics (QCD), I will explain the "sign problem". This problem cannot be solved by standard MC sampling methods and prohibits a numerical analysis of interesting regimes in QCD and other physical theories. For the remainder of the talk existing and novel approaches are presented with related problems to tackle the sign problem, in other words, to be able to sample from probability distributions with a complex contribution.

Back to Machine Learning for Physics and the Physics of Learning