Simple explanation of Landauer's bound and its ineffectiveness for multivalued logic

TL;DR

This paper critically examines Landauer's bound in multivalued logic, highlighting its limitations and complexities through thermodynamic and information-theoretic perspectives, including Szilard's demon and Galois connections.

Contribution

It provides a simplified explanation of Landauer's bound and demonstrates its ineffectiveness for multivalued logic systems, clarifying misconceptions and pitfalls.

Findings

Landauer's bound has limitations in multivalued logic.

Different thermodynamic memory implementations are analyzed.

Relation to Galois connection offers new insights.

Abstract

We discuss, using recent results on the Landauer's bound in multivalued logic, the difficulties and pitfalls of how to apply this principle. The presentation is based on Szilard's version of Maxwell's demon experiment and use of equilibrium Thermodynamics. Different versions of thermodynamical/mechanical memory are presented - one-hot encoding version and the implementation based on reversed Szilard's experiment. Relation of the Landauer's principle to Galois connection is explained in detail.