Abstract:``Nishimori line'' is a line or hypersurface in the parameter space of systems with quenched disorder, where simple expressions of the averages of physical quantities over the quenched random variables are obtained. It has been playing an important role in the theoretical studies of the random frustrated systems since its discovery around 1980. In this paper, a novel interpretation of the Nishimori line from the viewpoint of statistical information processing is presented. Our main aim is the reconstruction of the whole theory of the Nishimori line from the viewpoint of Bayesian statistics, or, almost equivalently, from the viewpoint of the theory of error-correcting codes. As a byproduct of our interpretation, counterparts of the Nishimori line in models without gauge invariance are given. We also discussed the issues on the ``finite temperature decoding'' of error-correcting codes in connection with our theme and clarify the role of gauge invariance in this topic.

What problem does this paper attempt to address?

The core problem that this paper attempts to solve is to provide a novel interpretation of the Nishimori line, especially to reconstruct the theory of the Nishimori line from the perspectives of Bayesian Statistics and coding theory. Specifically: 1. **Interpret the physical meaning of the Nishimori line**: The paper aims to clarify the physical meaning behind the Nishimori line and its proof motivation, which has not been fully clarified in previous research. 2. **Interpret the Nishimori line from the perspective of information processing**: By introducing the Bayesian statistical framework, the author links the Nishimori line with statistical information processing, shows the unexpected connection between two different fields, and reveals the significance of the techniques for deriving the Nishimori line. 3. **Generalize to models without gauge invariance**: The paper also discusses how to find concepts similar to the Nishimori line in the absence of gauge invariance. This provides a new perspective for studying other types of random systems. 4. **Finite - temperature decoding problem**: The paper explores issues related to finite - temperature decoding, especially its application in coding theory, and further clarifies the role of gauge invariance in this topic. ### Main contributions - **New interpretation**: The paper provides a novel interpretation of the Nishimori line, regarding it as a line for inference (or decoding) using the true probability structure of the generated data. This consistency greatly simplifies the calculation of the average values of various physical quantities. - **Generalize to other models**: The paper shows that even in models without gauge invariance, concepts similar to the Nishimori line can be defined, and corresponding energy identities, specific heat inequalities, etc. can be derived. - **Information processing perspective**: By introducing Bayesian statistics and coding theory, the paper reveals the potential applications of the Nishimori line in the information processing field, providing a bridge for interdisciplinary research. ### Mathematical formulas The mathematical formulas involved in the paper mainly include: - **Bayesian posterior distribution**: \[ p(x|y)=\frac{p(y|x)\pi(x)}{\sum_x p(y|x)\pi(x)} \] - **Energy function**: \[ -E_\alpha(x,y)=\alpha\sum_{(i,j)}y_{ij}x_ix_j \] - **Normalization factor**: \[ Z_\alpha=\sum_y\exp(-E_\alpha(x,y)) = (\exp(\alpha)+\exp(-\alpha))^M \] - **Nishimori line condition**: \[ q = \frac{\exp(-\alpha_0)}{\exp(\alpha_0)+\exp(-\alpha_0)} \] These formulas play a crucial role in understanding the properties of the Nishimori line and its generalization in different models.

The Nishimori line and Bayesian Statistics

Gibbs-Bogoliubov inequality on Nishimori line

Critical exponents at the Nishimori point

Griffiths Inequalities for Ising Spin Glasses on the Nishimori Line

Critical lines in the pure and disordered $O(N)$ model

Anomalous distribution of magnetization in an Ising spin glass with correlated disorder

Central limit theorem for the overlaps on the Nishimori line

About the de Almeida–Thouless line in neural networks

Dynamical critical behavior on the Nishimori point of frustrated Ising models

Toward mean-field bound for critical temperature on Nishimori line

Gauge theory for mixed $p$-spin glasses

About the de Almeida-Thouless line in neural networks

Bayesian Inference for Contemporary Lattice Quantum Field Theory

Bayesian Renormalization

Correlation versus Dissipation in a Non-Hermitian Anderson Impurity Model

Disorder Line and Incommensurate Floating Phases in the Quantum Ising Model on an Anisotropic Triangular Lattice

Bayesian Inference of General Noise Model Parameters from Surface Code's Syndrome Statistics

Statistical physical view of statistical inference in Bayesian linear regression model

Realizing the Nishimori transition across the error threshold for constant-depth quantum circuits

Information Bounds on phase transitions in disordered systems