Thermodynamic Realizability of Hidden Markov Processes: Attainment, Observable Certificates, and the Price of Architecture
Abstract
What is the least dissipative finite Markov machine that can reproduce a given stochastic process exactly? A basic compactness worry is that a minimizing sequence might lower its cost only by sending microscopic rates to infinity, never converging to an actual finite-rate machine. We prove that this escape is impossible at fixed hidden dimension. States of vanishing stationary occupation can be traced out, and infinitely fast conductance classes can be contracted, without increasing entropy production or losing the limiting observed path law. Hence the exact-law cost Vn(P) is attained and lower semicontinuous at every fixed state count, with no cap on rates or mean activity. Adding a positive price κ per used hidden state yields an attained optimum over all finite dimensions and an exact representation by globally calibrated lower certificates built from finitely many observable path expectations. The lower envelope Gκ(P)=minn[Vn(P)+κn] defines an architecture phase diagram whose slope is the selected state count; divergence of that count as κ↓0 is equivalent to failure of any finite machine to attain the all-dimension infimum V∞(P). For the two-block renewal family Pq, we derive complete finite-dimensional phase-type cancellation equations, give an explicit four-phase counterexample to a tempting restricted parameterization, and prove strictly positive minimal dissipation at every feasible finite dimension for all 0<q<1. These results reduce the remaining problem to one sharp question: can growing hidden architecture drive the dissipation of Pq to its all-dimension infimum without any finite optimizer, and in the strongest scenario, to zero?
Related
- Compositional Operational Rank, Smooth Realizability, and Dimensional Reduction
- Finite attainment and efficient computation of regularized quantum channel relative entropy
- Cosmic topology. Part Va. Information content of the observable Universe
- Dynamic Observable Subspaces: Fast Simulation of Observable Quantum Dynamics
- Learning When Black Hole Hair Is Observable: Physical Identifiability, Generalization, and Observable Complementarity
BibTeX
@misc{aznagulov2026thermodynamic,
title = {{Thermodynamic Realizability of Hidden Markov Processes: Attainment, Observable Certificates, and the Price of Architecture}},
author = {Murad Aznagulov},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.14623},
eprint = {2609.14623},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.14623},
}
RIS
TY - UNPB
TI - Thermodynamic Realizability of Hidden Markov Processes: Attainment, Observable Certificates, and the Price of Architecture
AU - Murad Aznagulov
PY - 2026
DA - 2026/09/13
PB - arXiv
DO - 10.48550/arxiv.2609.14623
UR - https://arxiv.org/abs/2609.14623
AB - What is the least dissipative finite Markov machine that can reproduce a given stochastic process exactly? A basic compactness worry is that a minimizing sequence might lower its cost only by sending microscopic rates to infinity, never converging to an actual finite-rate machine. We prove that this escape is impossible at fixed hidden dimension. States of vanishing stationary occupation can be traced out, and infinitely fast conductance classes can be contracted, without increasing entropy production or losing the limiting observed path law. Hence the exact-law cost V_n(P) is attained and lower semicontinuous at every fixed state count, with no cap on rates or mean activity. Adding a positive price κ per used hidden state yields an attained optimum over all finite dimensions and an exact representation by globally calibrated lower certificates built from finitely many observable path expectations. The lower envelope G_κ(P)=min_n[V_n(P)+κn] defines an architecture phase diagram whose slope is the selected state count; divergence of that count as κ↓0 is equivalent to failure of any finite machine to attain the all-dimension infimum V_∞(P). For the two-block renewal family P_q, we derive complete finite-dimensional phase-type cancellation equations, give an explicit four-phase counterexample to a tempting restricted parameterization, and prove strictly positive minimal dissipation at every feasible finite dimension for all 0<q<1. These results reduce the remaining problem to one sharp question: can growing hidden architecture drive the dissipation of P_q to its all-dimension infimum without any finite optimizer, and in the strongest scenario, to zero?
ER -