Learning force fields from stochastic trajectories

When monitoring the dynamics of microscopic systems, disentangling deterministic forces from thermal noise is challenging. Indeed, we show that there is an information-theoretic bound on the rate at which information about the force field can be extracted from a trajectory. We propose a practical method, Stochastic Force Inference, that optimally uses this information to approximate force fields. This technique readily permits the evaluation of out-of-equilibrium currents and entropy production with a limited amount of data.

Ref: arXiv:1809.09650