Information-theoretic lower bound on energy cost of stochastic computation

Abstract

Physical systems are often simulated using a stochastic computation where different final states result from identical initial states. Here, we derive the minimum energy cost of simulating a data sequence of a general physical system by stochastic computation. We show that the cost is proportional to the difference between two information-theoretic measures of complexity of the data—the statistical complexity and the predictive information . We derive the difference as the amount of information erased during the computation. Finally, we illustrate the physics of information by implementing the stochastic computation as a Gedanken experiment with a Szilard-type engine. The results create a new link between thermodynamics, information theory and complexity.