Information Geometry of the Gaussian Distribution in View of Stochastic Optimization

DOI10.1145/2725494.2725510MaRDI QIDQ2981895zbMATH OpenOpenAlexWikidataFDO

Authors Luigi Malagò, Giovanni Pistone

Publication date 10 May 2017

Published in Proceedings of the 2015 ACM Conference on Foundations of Genetic Algorithms XIII (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1145/2725494.2725510

zbMATH Keywords

exponential familyinformation geometrymultivariate Gaussian distributionstochastic natural gradient descentstochastic relaxation

Mathematics Subject Classification ID

Special Riemannian manifolds (Einstein, Sasakian, etc.) (53C25) Statistical aspects of information-theoretic topics (62B10) Stochastic programming (90C15) Measures of information, entropy (94A17)

Recommendations

• Optimization via Information Geometry
• Information geometry of the Gaussian space
• scientific article; zbMATH DE number 7426342
• Information geometry, simulation and complexity in Gaussian random fields
• Information geometry of \(q\)-Gaussian densities and behaviors of solutions to related diffusion equations
• Information geometry and asymptotic geodesics on the space of normal distributions
• Information geometry of smooth densities on the Gaussian space: Poincaré inequalities
• Information-geometric optimization algorithms: a unifying picture via invariance principles
• Information geometry of the power inverse Gaussian distribution
• scientific article; zbMATH DE number 3872359

Cites work

• Absorption time of the Moran process
• Approximating fixation probabilities in the generalized Moran process
• scientific article; zbMATH DE number 3148873 (Why is no real title available?)
• On the fixation probability of superstars

Cited in

(15)

• Wasserstein Riemannian geometry of Gaussian densities
• Theoretical foundation for CMA-ES from information geometry perspective
• Second-order optimization over the multivariate Gaussian distribution
• Regularized Aggregation of One-Off Probability Predictions
• Gradient flow of the stochastic relaxation on a generic exponential family
• scientific article; zbMATH DE number 7426342 (Why is no real title available?)
• Optimization via Information Geometry
• Objective improvement in information-geometric optimization
• Information-geometric optimization algorithms: a unifying picture via invariance principles
• Robust estimation of natural gradient in optimization by regularized linear regression
• Relations between entropy rate, entropy production and information geometry in linear stochastic systems
• Bayesian experimental design without posterior calculations: an adversarial approach
• Covariance-modulated optimal transport and gradient flows
• Learning symmetries and non-Euclidean data representations via collective dynamics of generalized Kuramoto oscillators
• Minimax estimation of partially-observed vector autoregressions

This page was built for publication: Information Geometry of the Gaussian Distribution in View of Stochastic Optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2981895)