arXiv research

Study extends bounds on sample covariance matrices with general dependence.

Researchers construct explicit bundles for ALF metrics, revealing rational patching matrices for gravitational instantons.

New combinatorial framework for geometric realizations of subword complexes.

New framework uses symmetry-based matrices for efficient, flexible NNs.

This work compresses heavy-tailed weight matrices for tighter generalization bounds.

Lower bounds on private estimation of Gaussian covariance matrices.

FedSPDnet improves federated learning for SPD matrices, outperforming existing methods.

Study on rotating surfaces in 4D space with matrices.

New estimator learns symmetric dynamics from few observations.

TensorGuide improves LoRA efficiency and expressivity through joint tensor-train optimization.

New metric tensor field on symmetric matrices simplifies eigenvector computation.

Estimates matrix trace optimization with statistical learning theory.

For any prime power $q$ and any dimension $s \ge 1$, we present a construction of $(t,s)$-sequences in base $q$ with finite-row generating matrices such that, for fixed $q$, the quality parameter $t$ is asymptotically optimal as a function of $s$ as $s \to \infty$. This is the first construction of $(t,s)$-sequences th…

A widespread approach in machine learning to evaluate the quality of a classifier is to cross -- classify predicted and actual decision classes in a confusion matrix, also called error matrix. A classification tool which does not assume distributional parameters but only information contained in the data is based on th…

Like most learning algorithms, the multilayer perceptrons (MLP) is designed to learn a vector of parameters from data. However, in certain scenarios we are interested in learning structured parameters (predictions) in the form of symmetric positive definite matrices. Here, we introduce a variant of the MLP, referred to…

The paper models financial correlation matrices using permutation invariant Gaussian models and predicts market anomalies.

Develops a novel stochastic algorithm for diagonal estimation of large matrices.

A method learns matrix factorization from diverse matrices and applies the knowledge to unseen matrices.

A parsimonious model reduces over-parameterization in skewed matrix variate mixtures.

The Riemannian Bures metric on the space of (normalized) complex positive matrices is used for parameter estimation of mixed quantum states based on repeated measurements just as the Fisher information in classical statistics. It appears also in the concept of purifications of mixed states in quantum physics. Here we d…

We consider the problem of sampling from posterior distributions for Bayesian models where some parameters are restricted to be orthogonal matrices. Such matrices are sometimes used in neural networks models for reasons of regularization and stabilization of training procedures, and also can parameterize matrices of bo…

Paper develops new method for detecting latent structure in large symmetric data matrices.

A new model uses Toeplitz matrices to analyze time-series data transitions.

Kalman filtering and smoothing algorithms are used in many areas, including tracking and navigation, medical applications, and financial trend filtering. One of the basic assumptions required to apply the Kalman smoothing framework is that error covariance matrices are known and given. In this paper, we study a general…

This work proves the asymptotic freeness of layerwise Jacobians in MLPs with Haar orthogonal matrices.

The density matrices are positively semi-definite Hermitian matrices of unit trace that describe the state of a quantum system. The goal of the paper is to develop minimax lower bounds on error rates of estimation of low rank density matrices in trace regression models used in quantum state tomography (in particular, i…

A new matrix concentration inequality for random products of matrices.

Kaleidoscope matrices improve model quality and inference speed.

This work considers a computationally and statistically efficient parameter estimation method for a wide class of latent variable models---including Gaussian mixture models, hidden Markov models, and latent Dirichlet allocation---which exploits a certain tensor structure in their low-order observable moments (typically…

Connections between nodes of fully connected neural networks are usually represented by weight matrices. In this article, functional transfer matrices are introduced as alternatives to the weight matrices: Instead of using real weights, a functional transfer matrix uses real functions with trainable parameters to repre…

The paper reduces the complexity of financial market correlation matrices to a 2x2 matrix.

The rebmix package provides R functions for random univariate and multivariate finite mixture model generation, estimation, clustering and classification. The paper is focused on multivariate normal mixture models with unrestricted variance-covariance matrices. The objective is to show how to generate datasets for a kn…

Optimizing over the set of orthogonal matrices is a central component in problems like sparse-PCA or tensor decomposition. Unfortunately, such optimization is hard since simple operations on orthogonal matrices easily break orthogonality, and correcting orthogonality usually costs a large amount of computation. Here we…

In the paper, we consider the problem of link prediction in time-evolving graphs. We assume that certain graph features, such as the node degree, follow a vector autoregressive (VAR) model and we propose to use this information to improve the accuracy of prediction. Our strategy involves a joint optimization procedure …

A connection is made between the Krammer representation and the Birman-Murakami-Wenzl algebra. Inspired by a dimension argument, a basis is found for a certain irrep of the algebra, and relations which generate the matrices are found. Following a rescaling and change of parameters, the matrices are found to be identica…

We introduce a general framework for estimation of inverse covariance, or precision, matrices from heterogeneous populations. The proposed framework uses a Laplacian shrinkage penalty to encourage similarity among estimates from disparate, but related, subpopulations, while allowing for differences among matrices. We p…

New method for NMF without tuning parameter.

The low displacement rank (LDR) framework for structured matrices represents a matrix through two displacement operators and a low-rank residual. Existing use of LDR matrices in deep learning has applied fixed displacement operators encoding forms of shift invariance akin to convolutions. We introduce a class of LDR ma…

Many matching, tracking, sorting, and ranking problems require probabilistic reasoning about possible permutations, a set that grows factorially with dimension. Combinatorial optimization algorithms may enable efficient point estimation, but fully Bayesian inference poses a severe challenge in this high-dimensional, di…

Graph alignment problem solved with convex relaxations for correlated matrices.

AI-driven framework optimizes MCMC-based preconditioners for faster linear system solving.

We introduce a variational Bayesian neural network where the parameters are governed via a probability distribution on random matrices. Specifically, we employ a matrix variate Gaussian \cite{gupta1999matrix} parameter posterior distribution where we explicitly model the covariance among the input and output dimensions…

Muon optimizer outperforms GD in neural networks.

We introduce a stochastic process with Wishart marginals: the generalised Wishart process (GWP). It is a collection of positive semi-definite random matrices indexed by any arbitrary dependent variable. We use it to model dynamic (e.g. time varying) covariance matrices. Unlike existing models, it can capture a diverse …

DDD reformulated for sparse matrices, integrating trajectory and snapshot time series data.

New framework finds more efficient linear layers over structured matrices.

Autoencoder performance is predicted by eigenvalues of weight matrices.

Given a positive and unitarily invariant Lagrangian L defined in the algebra of Hermitian matrices, and a fixed interval $[a,b]\subset\mathbb R$, we study the action defined in the Lie group of $n\times n$ unitary matrices $\mathcal{U}(n)$ by $S(α)=\int_a^b L(\dotα(t))\,dt\,,$ where $α:[a,b]\to\mathcal{U}(n)$ is a …