Thermodynamics of statistical inference by cells

Abstract

The deep connection between thermodynamics, computation, and information is now well established both theoretically and experimentally. Here, we extend these ideas to show that thermodynamics also places fundamental constraints on statistical estimation and learning. To do so, we investigate the constraints placed by (nonequilibrium) thermodynamics on the ability of biochemical signaling networks within cells to estimate the concentration of an external signal. We show that accuracy is limited by energy consumption, suggesting that there are fundamental thermodynamic constraints on statistical inference.

15 pages, 5 figures. V2 includes edits and 2 new figures. V3 is updated to version published in Phys. Rev. Lett

References in corpus (3)

• Work and information processing in a solvable model of Maxwell's demon
• The thermodynamics of prediction
• Information-theoretic vs. thermodynamic entropy production in autonomous sensory networks

Cited by in corpus (10)

• Thermodynamic uncertainty relation for biomolecular processes
• Cost and Precision of Brownian Clocks
• Maxwell's demon in biochemical signal transduction with feedback loop
• Nonequilibrium sensing and its analogy to kinetic proofreading
• Stochastic Thermodynamics of Learning
• Limits of Predictions in Thermodynamic Systems: A Review
• Thermodynamic efficiency of learning a rule in neural networks
• Minimal informational requirements for fitness
• Limitations on concentration measurements and gradient discerning times in cellular systems
• Individual Sensing can Gain more Fitness than its Information