Unifying Three Perspectives on Information Processing in Stochastic Thermodynamics

TL;DR

This paper unifies three perspectives on information processing in stochastic thermodynamics, deriving generalized second laws that incorporate mutual information and Shannon entropy differences, thus extending the framework to include information reservoirs.

Contribution

It introduces a unified approach to stochastic thermodynamics that encompasses measurement, control, and information reservoirs, deriving generalized second laws from a single fluctuation theorem.

Findings

Derived generalized second laws involving mutual information and Shannon entropy differences.

Showed the relationship between different entropy production terms in feedback systems.

Extended stochastic thermodynamics to include information reservoirs like tapes.

Abstract

So far, feedback-driven systems have been discussed using (i) measurement and control, (ii) a tape interacting with a system or (iii) by identifying an implicit Maxwell demon in steady state transport. We derive the corresponding second laws from one master fluctuation theorem and discuss their relationship. In particular, we show that both the entropy production involving mutual information between system and controller and the one involving a Shannon entropy difference of an information reservoir like a tape carry an extra term different from the usual current times affinity. We thus generalize stochastic thermodynamics to the presence of an information reservoir.