# Biochemical Szilard engines for memory-limited inference

## TL;DR

This paper explores how minimal molecular Szilard engines with limited memory can extract work from complex, correlated environments, revealing the trade-offs between inference complexity, energy extraction, and robustness.

## Contribution

It introduces a molecular Szilard engine model for minimal inference, analyzing how environment complexity influences energy extraction and robustness considerations.

## Key findings

- More complex environments enable more effective energy extraction.
- Limited memory constrains the inference strategies and energy gains.
- Robustness against fluctuations requires additional energy reserves.

## 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.