CSCS Seminar | Tensor Networks for Chemical Reaction Network Rate Calculations

Coffee and snacks will be provided. This talk will be recorded for later viewing.

Abstract: In studying the dynamics of stochastic complex systems, it is common to employ Monte Carlo sampling of trajectories because it is typically impractical to solve for the evolution of the probability distribution over exponentially large state spaces. In my talk, I will present a case study (PRX, 13, 041006, 2023) showing a numerically tractable route to studying the stochastic dynamics without sampling. The illustrative example involves a "rare-event rate calculation" which extracts the timescale to exhibit a stochastic switch between bistable states. The talk will provide a primer on how tensor networks can be employed to perform numerically controllable stochastic many-body dynamics with an eye toward demystifying how similar approaches can be employed in other domains.