Hamilton–Jacobi equations for inference of matrix tensor products

نویسندگان

چکیده

We study the high-dimensional limit of free energy associated with inference problem finite-rank matrix tensor products. In general, we bound from above by unique solution to a certain Hamilton-Jacobi equation. Under additional assumptions on nonlinearity in equation which is determined explicitly model, identify solution. Two notions solutions, weak solutions and viscosity are considered, each has its own advantages requires different treatments. For concreteness, apply our results model i.i.d. entries symmetric interactions. particular, for first order even products, obtain estimates convergence rates; other odd orders, upper bounds obtained.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Recognising Tensor Products of Matrix Groups

As a contribution to the project for recognising matrix groups deened over nite elds, we describe an algorithm for deciding whether or not the natural module for such a matrix group can be decomposed into a non-trivial tensor product. In the aarmative case, a tensor decomposition is returned. As one component, we develop algorithms to compute p-local subgroups of a matrix group.

متن کامل

On Kronecker products, tensor products and matrix differential calculus

The algebra of the Kronecker products of matrices is recapitulated using a notation that reveals the tensor structures of the matrices. It is claimed that many of the difficulties that are encountered in working with the algebra can be alleviated by paying close attention to the indices that are concealed beneath the conventional matrix notation. The vectorisation operations and the commutation...

متن کامل

Joint Matrix-Tensor Factorization for Knowledge Base Inference

While several matrix factorization (MF) and tensor factorization (TF) models have been proposed for knowledge base (KB) inference, they have rarely been compared across various datasets. Is there a single model that performs well across datasets? If not, what characteristics of a dataset determine the performance of MF and TF models? Is there a joint TF+MF model that performs robustly on all da...

متن کامل

A numerical algorithm for solving a class of matrix equations

In this paper, we present a numerical algorithm for solving matrix equations $(A otimes B)X = F$  by extending the well-known Gaussian elimination for $Ax = b$. The proposed algorithm has a high computational efficiency. Two numerical examples are provided to show the effectiveness of the proposed algorithm.

متن کامل

Using operational matrix for numerical solution of fractional differential equations

In this article, we have discussed a new application of modification of hat functions on nonlinear multi-order fractional differential equations. The operational matrix of fractional integration is derived and used to transform the main equation to a system of algebraic equations. The method provides the solution in the form of a rapidly convergent series. Furthermore, error analysis of the pro...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Annales de l'I.H.P

سال: 2022

ISSN: ['0246-0203', '1778-7017']

DOI: https://doi.org/10.1214/21-aihp1183