Searcharxiv⌕ Search

arXiv subjects

Hong-Bin Chen

Publications and source records attributed to Hong-Bin Chen.

At least 19 recordsLinked to original sources

On Hamilton-Jacobi equations and perceptron models

The solution to a finite-dimensional Hamilton-Jacobi equation is classically known to admit a variational representation in two situations: when the nonlinearity is convex or concave, by the Hopf-Lax formula; and when the initial condition is convex or concave, by the Hopf formula. We identify a third situation: the space variable is split into two groups of coordinates, the nonlinearity is the scalar product of the two corresponding partial gradients, and the initial condition is the sum of a convex function of the first group of coordinates and of an arbitrary Lipschitz function of the second group. This structure is precisely that of the Hamilton-Jacobi equations describing the limit free energy of perceptron models. We illustrate this point by identifying the limit free energy of a class of perceptron models that can be rewritten as bipartite spin glasses, with Ising spins in one layer and bounded soft spins in the other. We also discuss how to extend the approach to several other classes of models, including general Ising and spherical perceptrons, possibly with labelled patterns.

math.PR↗

Parisi formula and large deviations for mean-field spin glasses with Mattis interactions

We consider convex mean-field vector spin glasses perturbed by a general Mattis interaction $N G(m_N)$, where $m_N$ is a bounded vector-valued empirical average and $G$ is continuous. We prove a Parisi-type variational formula for the limiting quenched free energy and a quenched large deviation principle for $m_N$, both at every temperature. The approach is simple and modular: the only model-specific spin-glass input is the variational formula for every linear tilt of $m_N$ together with differentiability of the resulting free energy. The classical Gärtner--Ellis, Varadhan, and Bryc transfers then treat an arbitrary continuous $G$. Thus the same argument applies to a broad class of empirical observables and Mattis potentials, without a separate constrained-free-energy calculation for each model. As a scalar specialization, the framework contains the Gaussian spin-glass surrogate used in mismatched rank-one matrix estimation.

math.PR↗

High-temperature free energy for non-convex mean-field spin glasses with Mattis interactions

We study enriched mean-field vector spin glasses whose covariance function need not be convex. In a quantitative short-time regime, corresponding to sufficiently high temperature in the usual physical parameterization, we prove convergence of the quenched free energy to the differentiable solution of a infinite-dimensional Hamilton--Jacobi equation proposed for non-convex models. We identify that solution through its characteristics and a unique fixed point. The argument does not assume replica symmetry. We then allow a general continuous Mattis interaction $N G(m_N)$, obtain the corresponding variational formula, and prove a quenched large deviation principle for $m_N$. As an explicit application, we treat a bipartite model with Mattis fields, which is closely related to the Restricted Boltzmann Machine. The application requires neither a replica-symmetry assumption nor ellipticity or Nishimori identities; its restriction is the same quantitative high-temperature condition as in the general theorem.

math.PR↗

Free energy of the Hopfield model

We identify the limit of the free energy of the Hopfield model with independent symmetric Bernoulli patterns, for every temperature and loading ratio. A Hubbard-Stratonovich transformation reduces the problem to that of identifying the limit free energy of a bipartite spin glass with $\pm 1$ and spherical spins. One key difficulty is that the patterns, which become the coupling coefficients in the bipartite model, are Bernoulli instead of Gaussian, and the limit free energy is sensitive to the law of these coefficients. We show that the model is in fact equivalent to one involving Gaussian patterns and also subject to a finite number of Bernoulli random fields. We then solve this model by adapting the techniques recently introduced to identify the limit free energies of spin glasses with either $\pm 1$ or spherical spins.

math.PR↗

Free energy of spherical vector spin glasses

We identify the limit free energy of spherical vector spin glasses by a variational formula. The covariance function need not be convex. Using the global symmetry of the spin configurations, we introduce a family of constraint measures that gives a disintegration of the original spin distribution. We first identify the limit free energy of each constrained model and then apply the Laplace method to recover that of the original model.

math.PR↗

Free energy of 2-spin spherical vector spin glasses

We compute the limit of free energy of 2-spin spherical vector spin glasses with possibly non-convex interaction and couplings satisfying a finite fourth-moment assumption. The couplings need not be Gaussian. The limit is given by a logarithmic-determinant variational formula, equivalently by a finite-dimensional matrix Dyson equation, and its zero-temperature limit yields the ground-state energy. The limiting formulas depend only on the covariance of the couplings. We also study fluctuations: at an interior minimizer, an explicit variance bound transfers the limiting law of a logarithmic linear statistic to the free energy. For Gaussian couplings, this yields Gaussian fluctuations. At a boundary minimizer, assuming a simple regular edge with GOE Tracy-Widom fluctuations and a one-edge reduction, we transfer the edge fluctuations to the free energy.

math.PR↗

A formula for the free energy of non-convex vector spin glasses

We consider vector mean-field spin glass models with independent and identically distributed spins without assuming the covariance function is convex. We identify the limit free energy as the minimum of a functional that can be written as the sum of a relative entropy and the solution of a Hamilton--Jacobi equation. The proof of this result relies on the observation that when the spin configurations of the system are constrained to have a given empirical measure, the corresponding free energy approximately satisfies a Hamilton--Jacobi equation with a uniformly bounded error.

math.PR↗

Noise-enhanced quantum kernels on analog quantum computers for estimating the non-Markovianity from sparse temporal data

The quantum kernel method, a promising quantum machine learning algorithm, possesses substantial potential for demonstrating a quantum advantage. While most quantum kernels are constructed in the context of gate-based quantum circuits, inspired by the idea of analog quantum computing with Rydberg atoms, here we construct an analog quantum kernel and a hybrid quantum kernel, and demonstrate their competitiveness against other standard baselines in the physically motivated task of estimating the non-Markovianity from sparse temporal data. Crucially, we report a counterintuitive phenomenon where the operational noise can enhance their performance when interatomic distances are appropriately tuned. Our analysis reveals that the noise can help the analog and hybrid quantum models keep the test MSE under control despite closely fitting the training data, yielding models with higher complexity and a training MSE lower than or comparable to their ideal counterparts. Our results not only reveal that the noise can counterintuitively benefit analog and hybrid quantum kernel methods, but also provide an efficient route to estimating the non-Markovianity with alleviated experimental demands.

quant-ph↗

Free energy of non-convex multi-species spherical spin glasses

We identify the limit free energy of all mean-field multi-species spin glasses defined on products of spheres. In particular, we do not make any convexity assumption on the covariance function of the Hamiltonian, and we allow for the presence of external fields. We also obtain a one-species reduction of the formula in the special case of balanced models.

math.PR↗

Free energy of multi-layer generalized linear models

We compute the high-dimensional limit of the free energy associated with a multi-layer generalized linear model. Under certain technical assumptions, we identify the limit in terms of a variational formula. The approach is to first show that the limit is a solution to a Hamilton-Jacobi equation whose initial condition is related to the limiting free energy of a model with one fewer layer. Then, we conclude with an iteration.

math.PR↗

Multi-point functions of a full-plane two-state fuzzy $4$-Potts model

We study the full-plane two-state fuzzy $4$-Potts model, obtained by assigning independent balanced $\{\pm1\}$ spins to the open clusters of a critical $q=4$ random-cluster configuration. Its spin field has the same law as a single-spin projection and as the polarization field of the isotropic Ashkin--Teller model at its Potts point. We prove that, after proper normalization, all even multi-point spin correlation functions converge to explicit conformally covariant Coulomb-gas type neutral charge sums. As a consequence, we prove convergence in law of the rescaled magnetization field and identify its limit in law with $2\mathopen{:}\cos(X/2)\mathclose{:}$ for a full-plane Gaussian free field $X$. The proof combines the Baxter--Kelland--Wu coupling, convergence of the six-vertex height function to the Gaussian free field, and a charge-completion mechanism: an enlarged discrete sum over charge assignments with total charge in $4\mathbb Z$ produces a combinatorial cancellation of connection patterns, while only the neutral charge sector survives in the scaling limit.

math.PR↗

FRSB in the SK spin glass: convergence to full-interval support at zero temperature

We prove full replica symmetry breaking for the zero-field Sherrington-Kirkpatrick model at zero temperature: the Parisi minimizer is absolutely continuous, has a smooth density, and has support $[0,1)$. At inverse temperature $β>1$, [arxiv.org/abs/2607.11756v3] recently proved that the Parisi measure has support $[0,q_β]$. Here, we show $q_β$ converges to $1$ as $β\to\infty$.

math.PR↗

One-sided large deviations for the ground-state energy of spin glasses

We describe the large deviations above its typical value of the maximal energy of a spin glass with +/-1 spins. Thanks to the relatively explicit description of the rate function we identify, we then show that the latter is asymptotically quadratic near its minimum if and only if an external magnetic field is present. The proof starts from a Parisi-type formula for the fractional moments of the partition function, which we then leverage to obtain the limit of the Laplace transform of the maximum energy. Using convex-duality arguments, we then rewrite this Laplace transform as a supremum over martingales, and thereby deduce the large-deviation principle with explicit rate function.

math.PR↗

On the free energy of vector spin glasses with non-convex interactions

The limit free energy of spin-glass models with convex interactions can be represented as a variational problem involving an explicit functional. Models with non-convex interactions are much less well-understood, and simple variational formulas involving the same functional are known to be invalid in general. We show here that a slightly weaker property of the limit free energy does extend to non-convex models. Indeed, under the assumption that the limit free energy exists, we show that this limit can always be represented as a critical value of the said functional. Up to a small perturbation of the parameters defining the model, we also show that any subsequential limit of the law of the overlap matrix is a critical point of this functional. We believe that these results capture the fundamental conclusions of the non-rigorous replica method.

math.PR↗

Color symmetry and ferromagnetism in Potts spin glass

We consider the Potts spin glass with additional ferromagnetic interaction parametrized by $t$. It has long been observed that the Potts color symmetry breaking for the spin glass order parameter is closely related to the ferromagnetic phase transition. To clarify this, we identify a single critical value $t_\mathrm{c}$, which marks the onset of both color symmetry breaking and the transition to ferromagnetism.

cond-mat.dis-nn↗

Simultaneous replica-symmetry breaking for vector spin glasses

We consider mean-field vector spin glasses with possibly non-convex interactions. Up to a small perturbation of the parameters defining the model, the asymptotic behavior of the Gibbs measure is described in terms of a critical point of an explicit functional. In this paper, we study some properties of these critical points. Under modest assumptions ensuring that different types of spins interact, we show that the replica-symmetry-breaking structures of the different types of spins are in one-to-one correspondence with one another. For instance, if some type of spins displays one level of replica-symmetry breaking, then so do all the other types of spins. This extends the recent results of [Electronic Journal of Probability, 27:1-75, 2022] and [Comm. Math. Phys., 394(3):1101-1152, 2022] that were obtained in the case of multi-species spherical spin glasses with convex interactions.

math.PR↗