Heat release by controlled continuous-time Markov jump processes

TL;DR

This paper derives equations for controlling continuous-time Markov jump processes to minimize heat release during state transitions, connecting discrete and continuous models and relating heat minimization to entropy production.

Contribution

It introduces a framework for optimal control of Markov jump processes to minimize heat, linking it to Hamilton-Jacobi equations and continuum limits.

Findings

Optimal control protocols derived for Markov jump processes.

Equivalence established between heat minimization and entropy production.

Continuum limit connects jump processes to diffusion models.

Abstract

We derive the equations governing the protocols minimizing the heat released by a continuous-time Markov jump process on a one-dimensional countable state space during a transition between assigned initial and final probability distributions in a finite time horizon. In particular, we identify the hypotheses on the transition rates under which the optimal control strategy and the probability distribution of the Markov jump problem obey a system of differential equations of Hamilton-Bellman-Jacobi-type. As the state-space mesh tends to zero, these equations converge to those satisfied by the diffusion process minimizing the heat released in the Langevin formulation of the same problem. We also show that in full analogy with the continuum case, heat minimization is equivalent to entropy production minimization. Thus, our results may be interpreted as a refined version of the second law of thermodynamics.