The thermodynamics of binary decisions: statistical inference in a two-state system
Year: 2026
Type: conference-paper
Abstract: We interpret first-order phase transitions as an inference problem in a two-state system. Using the maximum-entropy (MaxEnt) formalism, we derive the probability distribution for a two-phase material and for a formally equivalent binary-decision model inspired by a ball game. Relative stability is quantified by the log-probability ratio: latent heat corresponds to the energy difference between phases, and coexistence follows from equal probabilities, yielding a Clausius–Clapeyron-type relation for the coexistence line. More generally, the transition appears as an abrupt change in inferred state probabilities, showing that phase-transition phenomenology can emerge from constrained inference beyond its usual thermodynamic setting.
Authors: Diego Mora, Gonzalo Gutiérrez
Institution: University of Chile
Language: English
Subfield: Statistical and Nonlinear Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7162691345