Least squares estimation works well for symmetric positive semidefinite matrices without regularization.
problem Estimation of symmetric positive semidefinite matrices without regularization.
method Simple least squares estimation with extsf{spd} constraint.
result Constrained least squares estimation performs as well as regularization-based approaches.
This article provides the mathematical foundation for stochastically continuous affine processes on the cone of positive semidefinite symmetric matrices. This analysis has been motivated by a large and growing use of matrix-valued affine processes in finance, including multi-asset option pricing with stochastic volatil…
Denise learns a function to quickly decompose covariance matrices robustly.
problem Robustly decomposing covariance matrices for feature extraction.
method Deep learning for symmetric positive semidefinite matrices.
result Denise achieves state-of-the-art performance in decomposition quality and speed.
New inequalities for matrix supermartingales converge under various conditions.
problem Convergence and maximal inequalities of supermartingales in positive semidefinite matrices.
method Developed new concentration inequalities for matrix supermartingales.
result New inequalities for matrix supermartingales under different tail conditions.
Optimal dictionaries minimize the average squared error in representing random vectors.
problem Finding optimal dictionaries for minimizing ℓ2-norm of coefficients in random vector representations. method Using rank-1 decompositions of symmetric positive semidefinite matrices, explicit descriptions and polynomial-time algorithms for ℓ2-optimal dictionaries are provided. result Explicit descriptions and polynomial-time algorithms for ℓ2-optimal dictionaries are provided. New geometric framework for positive semidefinite matrices of fixed rank.
problem Statistical analysis of positive semidefinite matrices of fixed rank.
method Introducing a manifold S(n,p)∗ with Riemannian geometry and Lie group structure. result Analytical closed forms for geodesics and Fréchet means.
New scalable geometric framework for SPD matrices.
problem Costly spectral computations in SPD matrix analysis.
method Efficient computation of extreme generalized eigenvalues through Hilbert and Thompson geometries of the semidefinite cone.
result Existence and uniqueness of a novel iterative mean of SPD matrices.
An algorithm for computing positive semidefinite factorizations of matrices.
problem Computing positive semidefinite factorizations of matrices.
method Non-commutative extension of Lee-Seung's algorithm (Matrix Multiplicative Update, MMU).
result The MMU algorithm ensures PSD updates and achieves critical points.
A new matrix concentration inequality for random products of matrices.
problem Understanding the behavior of random matrix products under bounded independent positive semidefinite matrices.
method Developed a non-asymptotic concentration inequality for the product of matrices.
result The inequality provides a bound on the deviation of the matrix product from its expected value.
Solves matrix completion for rectangular matrices using gradient descent.
problem Matrix completion for rectangular matrices with limited data.
method Lifts matrix to positive semidefinite, optimizes over semidefinite factor using gradient descent.
result Algorithm converges linearly to global optimum with high probability.
We present a hybrid algorithm for optimizing a convex, smooth function over the cone of positive semidefinite matrices. Our algorithm converges to the global optimal solution and can be used to solve general large-scale semidefinite programs and hence can be readily applied to a variety of machine learning problems. We…
New algorithm approximates large psd matrices from sketches.
problem Large-scale positive-semidefinite matrices from streaming data.
method Combines Nystrom approximation with rank truncation.
result Achieves prescribed relative error in Schatten 1-norm.
We consider a short rate model, driven by a stochastic process on the cone of positive semidefinite matrices. We derive sufficient conditions ensuring that the model replicates normal, inverse or humped yield curves.
Introduces a new model for mapping matrices to matrices, subsuming linear regression.
problem Learning matrix-to-matrix mappings from data.
method Partial trace regression model, leveraging quantum information theory.
result Relevance demonstrated in matrix-to-matrix regression and positive semidefinite matrix completion.
Study differential properties of symmetric real matrices with a specific metric.
problem Differential geometric properties of symmetric real matrices.
method Trace metric on non-singular symmetric real matrices, isometries of positive definite matrices.
result Description of the full group of isometries for positive definite matrices.
The SCMU algorithm computes cone factorizations for symmetric cones, improving upon existing methods.
problem Computing cone factorizations for symmetric cones in optimization.
method Introduces and analyzes the symmetric-cone multiplicative update (SCMU) algorithm.
result The SCMU algorithm non-decreases the squared loss objective.
Mirror descent algorithm recovers low-rank matrices in matrix sensing.
problem Matrix sensing with low-rank matrices under certain conditions.
method Discrete-time mirror descent applied to empirical risk with Bregman divergence analysis.
result Mirror descent converges to a matrix minimizing a specific nuclear norm-related quantity.
New PSDMF algorithms derived from PR and ARM methods.
problem Positive semidefinite matrix factorization (PSDMF) challenges.
method Design PSDMF algorithms based on phase retrieval (PR) and affine rank minimization (ARM) methods.
result New PSDMF algorithms inherit numerical properties from PR and ARM methods.
Proposes a new method for publishing covariance matrices while maintaining privacy and preserving matrix properties.
problem Publishing covariance matrices while ensuring differential privacy and maintaining positive semi-definiteness.
method Uses a Wishart distribution to generate matrix noise for differential privacy in principal component analysis.
result Demonstrates better utility compared to the Laplace mechanism and provides a near optimal bound.
The paper introduces new processes for modeling multivariate volatility.
problem Developing new stochastic processes for multivariate volatility modeling.
method Introducing Volterra Wishart and Volterra pure jump processes with fractional kernels.
result Affine covariance processes for multivariate volatility modeling.
This work improves understanding of symmetrizing Bregman divergences on positive definite matrices.
problem Understanding which mean to use for symmetrizing Bregman divergences on positive definite matrices.
method Axiomatic definition of mean functionals and variational principles over the cone of positive definite matrices.
result The arithmetic mean is canonical for forward symmetrization, and the arithmetic, log-Euclidean, and harmonic means for reverse symmetrization.
Improved method for computing Fréchet means on SPD matrices.
problem Computing Fréchet means on the manifold of SPD matrices.
method Random matrix theory-based approach for estimating Fréchet means.
result Significantly outperforms state-of-the-art methods in experiments.
Introduces matrix MLP for learning symmetric positive definite matrices.
problem Learning structured parameters like symmetric positive definite matrices.
method Develops matrix multilayer perceptron (matrix MLP) for structured parameter learning.
result Extends variational autoencoder (VAE) for dense covariance matrices.
Global stability bounds for matrix frames in phase retrieval problems.
problem Phase retrieval for matrix frames in various applications.
method Computable global stability bounds for the quasi-linear analysis map β, using Whitney stratification of positive semidefinite matrices of low rank.
result Novel conditions for a frame to be generalized phase retrievable.
This paper derives radial fields on manifolds of symmetric positive definite matrices.
problem Lack of an expression for radial fields on manifolds of symmetric positive definite matrices.
method Derives an expression for radial fields on manifolds of symmetric positive definite matrices.
result Derives an expression for radial fields on manifolds of symmetric positive definite matrices.
Proposes a new Sliced-Wasserstein distance for covariance matrices in M/EEG signals.
problem Efficiently dealing with distributions of covariance matrices in M/EEG multivariate time series.
method Defines a Sliced-Wasserstein distance for symmetric positive definite matrices and applies it to brain-age prediction and Brain Computer Interface applications.
result Demonstrates computational efficiency and strong theoretical guarantees for the proposed distance.
Improved covariance matrix estimation for portfolio optimization with guaranteed PSD and controlled conditioning.
problem Guaranteeing positive semidefinite ness and controlling spectral conditioning in IQ estimators.
method Introducing squeezing identity and atomic-IQ parameterization to construct structured channel matrices with PSD guarantees and analytic eigen floor for conditioning control.
result Atomic-IQ improves Sharpe ratios and delivers a more stable risk profile compared to standard estimators.
New method classifies manifold-valued data using Riemannian geometry.
problem Classifying data on curved Riemannian manifolds.
method Probabilistic Learning Vector Quantization on Symmetric Positive Definite Matrices.
result The method outperforms traditional Euclidean methods on manifold-valued data.
New Riemannian metric for SPD matrices avoids swelling effect.
problem Efficiency and stability in computing with SPD matrices.
method Log-Cholesky decomposition and Lie group structure.
result Log-Cholesky average maintains determinant bounds.
Algorithm finds positive braids with specific crossing matrices.
problem Finding positive braids with given crossing matrices.
method Finite algorithm using decomposition of braids.
result Algorithm determines existence of positive braids with specific matrices.
Paper introduces a generalized Bures-Wasserstein geometry for SPD matrices.
problem Understanding the geometry of SPD matrices for machine learning.
method Proposes a generalized Bures-Wasserstein geometry parameterized by a symmetric positive definite matrix.
result The GBW geometry outperforms the BW geometry in machine learning applications.
SDP helps recover hidden communities from symmetric data matrices.
problem Recovering hidden communities from symmetric data matrices.
method Semidefinite programming relaxation for maximum likelihood estimation.
result SDP achieves information-theoretic limits for large communities, but is suboptimal for smaller communities.
Motivated by some applications in signal processing and machine learning, we consider two convex optimization problems where, given a cone K, a norm ∥⋅∥ and a smooth convex function f, we want either 1) to minimize the norm over the intersection of the cone and a level set of f, or 2) to minimize over the…
Paper introduces a new distance measure for Gaussian Mixture Models.
problem Developing a new distance measure for Gaussian Mixture Models.
method Embedding K-component Gaussian Mixture Models into the manifold of symmetric positive definite matrices and calculating a lower bound for the Fisher-Rao metric.
result Demonstrated effectiveness through experiments on standard datasets.
The paper explores Cholesky decompositions for symmetric matrices and their geometric properties.
problem Understanding the structure and properties of symmetric matrices through Cholesky decompositions.
method Introducing cones of symmetric matrices, proving Cholesky-type factorizations, and showing geometric properties.
result Each symmetric matrix admits an uncountable family of Cholesky-type factorizations, and these cones are isometric Riemannian manifolds.
New framework solves low-rank optimization problems to certifiable optimality.
problem Low-rank optimization problems with certifiable solutions.
method Mixed-Projection Conic Optimization framework using symmetric projection matrices and outer-approximation algorithms.
result Solves low-rank problems to certifiable optimality, outperforming existing methods.
New algorithm for online optimization over symmetric cones, unifying previous methods.
problem Online convex optimization over symmetric cones.
method Symmetric-Cone Multiplicative Weights Update (SCMWU) algorithm.
result SCMWU is a no-regret algorithm.
New k-means method clusters radar image sequences using SPD matrices.
problem Clustering radar image sequences efficiently.
method Developed k-means on SPD matrices for non-Euclidean data. result Effective clustering of radar image sequences via SPD matrices.
Researchers find optimal dictionaries for minimizing average squared coefficients in random vector representations.
problem Finding optimal dictionaries for minimizing the average squared coefficients in random vector representations.
method Using rank-1 decompositions and majorization theory, the study provides a complete characterization of optimal dictionaries.
result Complete characterization of ℓ2-optimal dictionaries with polynomial time algorithms. Knots and 4-manifolds linked via matrix kinking.
problem Understanding equivalence of symmetric matrices and their implications.
method Isotopy and kinking moves on Goeritz matrices.
result Every nonsingular symmetric integer matrix is kink-equivalent to positive or negative-definite matrices.
Paper develops Riemannian geometry for SPSD matrices with DA applications.
problem Riemannian geometry of SPSD matrices for DA.
method Closed-form expressions, approximations of geodesic path, PT, canonical representation.
result Proposes an algorithm for DA with improved performance.
New balanced metrics introduced for SPD matrices, improving metric choice.
problem Lack of principles for choosing SPD matrix metrics.
method Introducing balanced metrics that relate existing metrics.
result Two new balanced metric families introduced: mixed-power-Euclidean and mixed-power-affine.
New methods for sketching non-PSD matrices improve regression and optimization tasks.
problem Efficiently handling non-PSD matrices in computations.
method Developed novel matrix sketching techniques for non-PSD and complex matrices.
result Improved performance in convex and non-convex optimization, regression, and vector-matrix-vector queries.
Rank-one measurements limit feasible sets for low-rank PSD matrices.
problem Feasibility of PSD matrices under rank-one measurements.
method Characterization of feasible sets for PSD matrices given rank-one projections.
result Radius of feasible sets determines singleton solution sets for low-rank matrices.
New nodal domain theorems for symmetric matrices via signed graphs.
problem Establish nodal domain theorems for symmetric matrices.
method Explore signed graph structure to define nodal domains for any function.
result Improved lower bound estimates for the number of strong nodal domains.
Study of strictly accretive matrices using Finsler geometry.
problem Characterize the set of strictly accretive matrices.
method Introduced Finsler metrics and characterized geodesics and distance.
result Geodesic distance applied to matrix approximation problem.
Researchers approximate partition functions on Riemannian spaces in the large N limit.
problem Computing normalization factors (partition functions) on Riemannian symmetric spaces is challenging.
method Approximation techniques in the large N limit, including saddle-point equations.
result Formulas for leading order terms in the large N limit of SPD matrices and related spaces.
Trace norm regularization is a popular method of multitask learning. We give excess risk bounds with explicit dependence on the number of tasks, the number of examples per task and properties of the data distribution. The bounds are independent of the dimension of the input space, which may be infinite as in the case o…