scientific article; zbMATH DE number 7306852

From MaRDI portal
Publication:5148924

MaRDI QIDQ5148924

James Martens

Publication date: 5 February 2021

Full work available at URL: https://arxiv.org/abs/1412.1193

Title: zbMATH Open Web Interface contents unavailable due to conflicting licenses.



Related Items (30)

A fully stochastic second-order trust region methodQNG: A Quasi-Natural Gradient Method for Large-Scale Statistical LearningInformation geometry of physics-informed statistical manifolds and its use in data assimilationSketch-based empirical natural gradient methods for deep learningModel-Centric Data Manifold: The Data Through the Eyes of the ModelA distributed optimisation framework combining natural gradient with Hessian-free for discriminative sequence trainingEpistemic uncertainty quantification in deep learning classification by the delta methodApproximate Newton Policy Gradient AlgorithmsSemi-implicit back propagationRobust federated learning under statistical heterogeneity via hessian-weighted aggregationAn overview of stochastic quasi-Newton methods for large-scale machine learningOn the locality of the natural gradient for learning in deep Bayesian networksInvariance properties of the natural gradient in overparametrised systemsEfficient Natural Gradient Descent Methods for Large-Scale PDE-Based Optimization ProblemsUnnamed ItemGeometry and convergence of natural policy gradient methodsDeep learning and geometric deep learning: An introduction for mathematicians and physicistsMulti-agent natural actor-critic reinforcement learning algorithmsThe limited-memory recursive variational Gaussian approximation (L-RVGA)Riemannian Natural Gradient MethodsDiscriminative Bayesian filtering lends momentum to the stochastic Newton method for minimizing log-convex functionsOptimization Methods for Large-Scale Machine LearningStochastic sub-sampled Newton method with variance reductionWarped Riemannian Metrics for Location-Scale ModelsVariational Bayes on manifoldsThe recursive variational Gaussian approximation (R-VGA)Laplace approximation and natural gradient for Gaussian process regression with heteroscedastic Student-\(t\) modelStructure-preserving deep learningUnderstanding approximate Fisher information for fast convergence of natural gradient descent in wide neural networks*Parametrisation independence of the natural gradient in overparametrised systems


Uses Software


Cites Work


This page was built for publication: