Thermodynamics of classifiers
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913157959647232 |
|---|---|
| author | Hasegawa, Yoshihiko |
| author_facet | Hasegawa, Yoshihiko |
| contents | The Landauer principle bridges the energetic cost and information processing, showing that irreversible computation inevitably demands energy dissipation. As energy demands from computation continue to rise, approximate computing has attracted considerable attention. Approximate computing is based on the idea that energy consumption can be reduced by sacrificing computational accuracy. This raises a fundamental question about the relationship between error and thermodynamic cost in information processing. In this study, we derive the error-cost trade-off in the binary classifier by considering classification based on Markov processes. We obtain the lower bounds on the Bayes error in terms of thermodynamic costs such as entropy production and dynamical activity. Our results show that when entropy production or dynamical activity vanishes, the Bayes error reaches $1/2$, equivalent to random guessing, while greater thermodynamic costs enable lower error. This establishes a fundamental trade-off between error and cost in information processing by thermodynamic systems. Because the Bayes error provides the lowest achievable error among all possible classifiers, the classification error cannot fall below the obtained bounds given the entropy production or dynamical activity. We also discuss the quantum generalization and show that the Bayes error of the quantum classifier is bounded from below by the variance of the Hamiltonian. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_24365 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Thermodynamics of classifiers Hasegawa, Yoshihiko Statistical Mechanics Quantum Physics The Landauer principle bridges the energetic cost and information processing, showing that irreversible computation inevitably demands energy dissipation. As energy demands from computation continue to rise, approximate computing has attracted considerable attention. Approximate computing is based on the idea that energy consumption can be reduced by sacrificing computational accuracy. This raises a fundamental question about the relationship between error and thermodynamic cost in information processing. In this study, we derive the error-cost trade-off in the binary classifier by considering classification based on Markov processes. We obtain the lower bounds on the Bayes error in terms of thermodynamic costs such as entropy production and dynamical activity. Our results show that when entropy production or dynamical activity vanishes, the Bayes error reaches $1/2$, equivalent to random guessing, while greater thermodynamic costs enable lower error. This establishes a fundamental trade-off between error and cost in information processing by thermodynamic systems. Because the Bayes error provides the lowest achievable error among all possible classifiers, the classification error cannot fall below the obtained bounds given the entropy production or dynamical activity. We also discuss the quantum generalization and show that the Bayes error of the quantum classifier is bounded from below by the variance of the Hamiltonian. |
| title | Thermodynamics of classifiers |
| topic | Statistical Mechanics Quantum Physics |
| url | https://arxiv.org/abs/2605.24365 |