Abstract:We study the law of a random field $f_N(\boldsymbol{\sigma})$ evaluated at a random sample from the Gibbs measure associated to a Gaussian field $H_N(\boldsymbol{\sigma})$. In the high-temperature regime, we show that bounds on the probability that $f_N(\boldsymbol{\sigma})\in A$ for $\boldsymbol{\sigma}$ randomly sampled from the Gibbs measure can be deduced from similar bounds for deterministic $\boldsymbol{\sigma}$ under the conditional Gaussian law given that $H_N(\boldsymbol{\sigma})/N=E$ for $E$ close to the derivative $F'(\beta)$ of the free energy (which is the typical value of $H_N(\boldsymbol{\sigma})/N$ under the Gibbs measure). In the more challenging low-temperature regime we restrict to $k$-RSB spherical spin glasses, proving a similar result, now with a more elaborate conditioning. Namely, with $q_i$ denoting the locations of the non-zero atoms of the Parisi measure, in addition to specifying that $H_N(\boldsymbol{\sigma})/N=E$, here one needs to also condition on the energy and its gradient at points $\mathbf{x}_1,\ldots,\mathbf{x}_k$ such that $\langle \mathbf{x}_i,\mathbf{x}_j\rangle/N=q_{i\wedge j}$ and $\langle \mathbf{x}_i,\boldsymbol{\sigma}\rangle/N\approx q_{i}$. Like in the high-temperature phase, the energy and gradient values on which one conditions are also specified by the model's Parisi measure. As an application, we compute the Franz-Parisi potential at any temperature. Our results are also relevant to the study of Langevin dynamics with initial conditions distributed according to the Gibbs measure, which is one of the main motivations of this work.

What problem does this paper attempt to address?

The problems that this paper attempts to solve mainly focus on two aspects: 1. **High - temperature phase**: Under high - temperature (small β value) conditions, the asymptotic behavior of the random field $ f_N(\sigma) $ on samples $ \sigma $ randomly drawn from the Gibbs measure associated with the Gaussian field $ H_N(\sigma) $ is studied. Specifically, the author proves that the probability bounds of $ f_N(\sigma) $ under the Gibbs measure can be derived through the conditional Gaussian distribution. 2. **Low - temperature phase**: Under low - temperature (large β value) conditions, the more complex $ k $-step replica - symmetry - breaking ($ k $-RSB) spherical spin - glass model is studied. The author proves that in this case, a more complex conditioning method is required, involving additional points $ x_1,\ldots,x_k $, which are conditioned to be critical points under a specific geometric structure, and the energy and gradient values of these points need to be specified. ### Main contributions - **High - temperature phase**: - The author proves that under high - temperature conditions, the behavior of $ f_N(\sigma) $ under the Gibbs measure can be described by the conditional Gaussian distribution. - See Theorem 1.1 for specific results. This theorem shows that if certain probability bounds under the conditional Gaussian distribution hold, then the corresponding probability bounds under the Gibbs measure also hold. - **Low - temperature phase**: - The author proposes a more complex conditioning method through the study of the $ k $-RSB spherical spin - glass model under low - temperature conditions. - See Theorem 1.4 for specific results. This theorem shows that by conditioning on specific critical points, the asymptotic behavior of $ f_N(\sigma) $ under the Gibbs measure can be derived. ### Applications - **Franz - Parisi potential**: The author uses the above results to calculate the Franz - Parisi potential at any temperature. - **Dynamics study**: These results are also used to study the Langevin dynamics with the initial condition obeying the Gibbs measure. ### Methods - **High - temperature phase**: - Using the properties of the conditional Gaussian distribution and combining with the concentration of free energy, the asymptotic behavior of $ f_N(\sigma) $ is derived. - The proof method involves techniques in random constraint satisfaction problems. - **Low - temperature phase**: - Using pure - state decomposition and ultrametric properties, the Gibbs measure is related to the critical points. - The average value on the critical points is expressed through the Kac - Rice formula and information on the complexity of the critical points. ### Conclusion Through detailed mathematical analysis, this paper provides a profound understanding of the asymptotic behavior of the random field $ f_N(\sigma) $ under the Gibbs measure under different temperature conditions. Especially in the low - temperature phase, through a complex conditioning method, the key problems in the $ k $-RSB spherical spin - glass model are solved. These results are not only of great significance to theoretical physics but also provide powerful tools for research in related fields.

Disordered Gibbs measures and Gaussian conditioning

Dynamics for spherical spin glasses: disorder dependent initial conditions

Gaussian concentration bound and Ensemble equivalence in generic quantum many-body systems including long-range interaction

The free energy of a quantum Sherrington-Kirkpatrick spin-glass model for weak disorder

Spectral Gap Estimates in Mean Field Spin Glasses

The Discrete Gaussian model, I. Renormalisation group flow at high temperature

The Discrete Gaussian model, II. Infinite-volume scaling limit at high temperature

Free Energy Difference Fluctuations in Short-Range Spin Glasses

Conditional Independence of 1D Gibbs States with Applications to Efficient Learning

Quantitative approximate independence for continuous mean field Gibbs measures

Sampling from Mean-Field Gibbs Measures via Diffusion Processes

Size of chaos for Gibbs measures of mean field interacting diffusions

Structural analysis of Gibbs states and metastates in short-range classical spin glasses: indecomposable metastates, dynamically-frozen states, and metasymmetry

Hypergraph Counting and Mixed $p$-Spin Glass Models under Replica Symmetry

Crisanti-Sommers formula and simultaneous symmetry breaking in multi-species spherical spin glasses

Phase diagram and large deviations in the free-energy of mean-field spin-glasses

Gibbs Measures with Multilinear Forms

Quantitative Disorder Effects in Low-Dimensional Spin Systems

On Long Range Ising Models with Random Boundary Conditions

Strong replica symmetry for high-dimensional disordered log-concave Gibbs measures