Wasserstein Riemannian geometry of Gaussian densities

DOI10.1007/s41884-018-0014-4zbMath1408.53058arXiv1801.09269OpenAlexW2901940252WikidataQ115371041 ScholiaQ115371041MaRDI QIDQ1713654

Luigi Malagò, Luigi Montrucchio, Giovanni Pistone

Publication date: 28 January 2019

Published in: Information Geometry (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1801.09269

zbMATH Keywords

Gaussian distributionnormal coordinatesinformation geometryRiemannian metricsWasserstein distanceRiemannian exponentialnatural gradientLevi-Civita covariant derivativeoptimization on positive-definite symmetric matrices

Mathematics Subject Classification ID

Geometric probability and stochastic geometry (60D05) Positive matrices and their generalizations; cones of matrices (15B48) Special Riemannian manifolds (Einstein, Sasakian, etc.) (53C25) Global geometric and topological methods (à la Gromov); differential geometric analysis on metric spaces (53C23)

Related Items

Riemannian Conjugate Gradient Methods: General Framework and Specific Algorithms with Convergence Analyses, When optimal transport meets information geometry, Entropy-regularized 2-Wasserstein distance between Gaussian measures, Bures–Wasserstein Minimizing Geodesics between Covariance Matrices of Different Ranks, Invariance properties of the natural gradient in overparametrised systems, \(O(n)\)-invariant Riemannian metrics on SPD matrices, Characterization of invariant inner products, Learning with symmetric positive definite matrices via generalized Bures-Wasserstein geometry, Affine statistical bundle modeled on a Gaussian Orlicz-Sobolev space, Geometry and convergence of natural policy gradient methods, Spatiotemporal analysis using Riemannian composition of diffusion operators, Designing universal causal deep learning models: The geometric (Hyper)transformer, The Schrödinger problem on the non-commutative Fisher-Rao space, Averaging Symmetric Positive-Definite Matrices, Riemannian Distances between Covariance Operators and Gaussian Processes, Optimal transport natural gradient for statistical manifolds with continuous sample space, Accelerated information gradient flow, Alpha Procrustes metrics between positive definite operators: a unifying formulation for the Bures-Wasserstein and Log-Euclidean/Log-Hilbert-Schmidt metrics, Hessian transport gradient flows, Bures-Wasserstein geometry for positive-definite Hermitian matrices and their trace-one subset

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Combinatorial optimization with information geometry: the Newton method
• Proper local scoring rules
• The distance between two random vectors wigh given dispersion matrices
• Wasserstein geometry of Gaussian measures
• Information geometry and its applications
• Some geometric calculations on Wasserstein space
• On the optimal mapping of distributions
• A class of Wasserstein metrics for probability distributions
• The Frechet distance between multivariate normal distributions
• A convexity principle for interacting gases
• A infinite-dimensional geometric structure on the space of all the probability measures equivalent to a given one
• On the Bures-Wasserstein distance between positive definite matrices
• Trail to a Lyapunov equation solver
• The Fritz John necessary optimality conditions in the presence of equality and inequality constraints
• THE GEOMETRY OF DISSIPATIVE EVOLUTION EQUATIONS: THE POROUS MEDIUM EQUATION
• Information Geometry of the Gaussian Distribution in View of Stochastic Optimization
• Computational Methods for Linear Matrix Equations
• On a Formula for the L2 Wasserstein Metric between Measures on Euclidean and Hilbert Spaces
• The Density Manifold and Configuration Space Quantization
• Polar factorization and monotone rearrangement of vector‐valued functions
• Kernel Density Estimation on Spaces of Gaussian Distributions and Symmetric Positive Definite Matrices
• Nonparametric Information Geometry
• Metric spaces, convexity and nonpositive curvature
• Riemannian geometry.
• Polar factorization of maps on Riemannian manifolds