Continuous information flow fluctuations

TL;DR

This paper derives fluctuation theorems for information flow and entropy production in feedback-controlled systems, providing insights into large deviations and thermodynamic fluctuations in small-scale nonequilibrium processes.

Contribution

It introduces new fluctuation theorems for information flow in driven diffusion processes and analyzes their implications for feedback cooling and Maxwell's demons.

Findings

Derived fluctuation theorems for information flow and entropy production.

Analyzed large deviation behaviors in feedback cooling models.

Extended results to arbitrary driven diffusion processes.

Abstract

Information plays a pivotal role in the thermodynamics of nonequilibrium processes with feedback. However, much remains to be learned about the nature of information fluctuations in small scale devices and their relation with fluctuations in other thermodynamics quantities, like heat and work. Here we derive a series of fluctuation theorems for information flow and partial entropy production in a Brownian particle model of feedback cooling and extend them to arbitrary driven diffusion processes. We then analyze the long-time behavior of the feedback-cooling model in detail. Our results provide insights into the structure and origin of large deviations of information and thermodynamic quantities in autonomous Maxwell's demons.