Maximum entropy principle for stationary states underpinned by stochastic thermodynamics
Ian J. Ford

TL;DR
This paper derives a maximum entropy principle for stationary states using stochastic thermodynamics, linking entropy production to system dynamics and recovering known equilibrium and nonequilibrium distributions.
Contribution
It introduces a dynamical basis for the maximum entropy principle by connecting entropy production components to stationary states in stochastic thermodynamics.
Findings
Recovers equilibrium probability density functions for conservative systems.
Derives stationary nonequilibrium distributions under nonisothermal conditions.
Explains thermodynamic anomalies between over- and underdamped dynamics.
Abstract
The selection of an equilibrium state by maximising the entropy of a system, subject to certain constraints, is often powerfully motivated as an exercise in logical inference, a procedure where conclusions are reached on the basis of incomplete information. But such a framework can be more compelling if it is underpinned by dynamical arguments, and we show how this can be provided by stochastic thermodynamics, where an explicit link is made between the production of entropy and the stochastic dynamics of a system coupled to an environment. The separation of entropy production into three components allows us to select a stationary state by maximising the change, averaged over all realisations of the motion, in the principal relaxational or nonadiabatic component, equivalent to requiring that this contribution to the entropy production should become time independent for all realisations.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
