Biochemical Szilard engines for memory-limited inference

Abstract

By developing and leveraging an explicit molecular realisation of a measurement-and-feedback-powered Szilard engine, we investigate the extraction of work from complex environments by minimal machines with finite capacity for memory and decision-making. Living systems perform inference to exploit complex structure, or correlations, in their environment, but the physical limits and underlying cost/benefit trade-offs involved in doing so remain unclear. To probe these questions, we consider a minimal model for a structured environment - a correlated sequence of molecules - and explore mechanisms based on extended Szilard engines for extracting the work stored in these non-equilibrium correlations. We consider systems limited to a single bit of memory making binary 'choices' at each step. We demonstrate that increasingly complex environments allow increasingly sophisticated inference strategies to extract more energy than simpler alternatives, and argue that optimal design of such machines should also consider the energy reserves required to ensure robustness against fluctuations due to mistakes.

References in corpus (28)

• Computational Mechanics: Pattern and Prediction, Structure and Simplicity
• Second law and Landauer principle far from equilibrium
• Ensemble and Trajectory Thermodynamics: A Brief Introduction
• On-chip Maxwell's demon as an information-powered refrigerator
• Work and information processing in a solvable model of Maxwell's demon
• Thermodynamics of a physical model implementing a Maxwell demon
• Nonequilibrium Thermodynamics of Chemical Reaction Networks: Wisdom from Stochastic Thermodynamics
• Thermodynamics with continuous information flow
• Stochastic thermodynamics of chemical reaction networks
• Efficiency of cellular information processing
• Imitating Chemical Motors with Optimal Information Motors
• An autonomous and reversible Maxwell's demon
• Thermodynamic efficiency of information and heat flow
• Finite-time erasing of information stored in fermionic bits
• Optimal Prediction by Cellular Signaling Networks
• Above and Beyond the Landauer Bound: Thermodynamics of Modularity
• Identifying Functional Thermodynamics in Autonomous Maxwellian Ratchets
• The thermodynamics of computational copying in biochemical systems
• Biochemical machines for the interconversion of mutual information and work
• Correlation-powered Information Engines and the Thermodynamics of Self-Correction
• Fundamental costs in the production and destruction of persistent polymer copies
• How an autonomous quantum Maxwell demon can harness correlated information
• Dependence of dissipation on the initial distribution over states
• Leveraging Environmental Correlations: The Thermodynamics of Requisite Variety
• Number of hidden states needed to physically implement a given conditional distribution
• A space-time tradeoff for implementing a function with master equation dynamics
• Physical Limitations of Work Extraction from Temporal Correlations
• Theory for the optimal detection of time-varying signals in cellular sensing systems

Cited by in corpus (5)

• Thermodynamic cost and benefit of memory
• Landauer Principle and Thermodynamics of Computation
• Provenance of Lyfe: Chemical Autonomous Agents Surviving through Associative Learning
• Dissipation reduction and information-to-measurement conversion in DNA pulling experiments with feedback protocols
• Simulation of reversible molecular mechanical logic gates and circuits