A Monte Carlo Time-Dependent Variational Principle

Abstract

We generalize the Time-Dependent Variational Principle (TDVP) to dissipative systems using Monte Carlo methods, allowing the application of existing variational classes for pure states, such as Matrix Product States (MPS), to the simulation of Lindblad master equation dynamics. The key step is to use sampling to approximately solve the Fokker-Planck equation derived from the Lindblad generators. An important computational advantage of this method, compared to other variational approaches to mixed state dynamics, is that it is "embarrassingly parallel".