Source-linked AI summary

The thermodynamics of prediction

Susanne Still, David A. Sivak, Anthony J. Bell, Gavin E. Crooks

arXiv:1203.3271v3cond-mat.stat-mechcs.ITq-bio.QM

TL;DR

The paper studies driven systems exposed to stochastic signals rather than fully known experiments. It demonstrates an equivalence between nonpredictive information retained by the system and dissipated energy, with predictive memory required for maximal energetic efficiency.

  • Problem

    Existing treatments typically assume the exact driving-signal time course is known, whereas realistic biological systems experience stochastic driving protocols.

  • Method

    The analysis models stochastic driving with a fixed system-dynamics kernel and compares conditional state distributions, excess work, and dissipation across protocols.

  • Results

    Instantaneous nonpredictive information is proportional to dissipated energy, while total nostalgia lower-bounds total average dissipation and excess work.

  • Takeaways & Limitations

    A memory-retaining computing device must be maximally predictive to approach Landauer’s limit and maximal energetic efficiency.

  • Takeaways & Limitations

    The system dynamics are assumed fixed for any given system, with no feedback from the system to the driving signal.

Abstract

from arXiv · show

A system responding to a stochastic driving signal can be interpreted as computing, by means of its dynamics, an implicit model of the environmental variables. The system's state retains information about past environmental fluctuations, and a fraction of this information is predictive of future ones. The remaining nonpredictive information reflects model complexity that does not improve predictive power, and thus represents the ineffectiveness of the model. We expose the fundamental equivalence between this model inefficiency and thermodynamic inefficiency, measured by dissipation. Our results hold arbitrarily far from thermodynamic equilibrium and are applicable to a wide range of systems, including biomolecular machines. They highlight a profound connection between the effective use of information and efficient thermodynamic operation: any system constructed to keep memory about its environment and to operate with maximal energetic efficiency has to be predictive.

Loading 1203.3271v3…