Evolutionary reinforcement learning of dynamical large deviations
Abstract
We show how to calculate the likelihood of dynamical large deviations using evolutionary reinforcement learning. An agent, a stochastic model, propagates a continuous-time Monte Carlo trajectory and receives a reward conditioned upon the values of certain path-extensive quantities. Evolution produces progressively fitter agents, eventually allowing the calculation of a piece of a large-deviation rate function for a particular model and path-extensive quantity. For models with small state spaces the evolutionary process acts directly on rates, and for models with large state spaces the process acts on the weights of a neural network that parameterizes the model's rates. This approach shows how path-extensive physics problems can be considered within a framework widely used in machine learning.
Cite
@article{arxiv.1909.00835,
title = {Evolutionary reinforcement learning of dynamical large deviations},
author = {Stephen Whitelam and Daniel Jacobson and Isaac Tamblyn},
journal= {arXiv preprint arXiv:1909.00835},
year = {2020}
}