Riemannian gradient methods for stochastic composition problems

DOI10.1016/J.NEUNET.2022.06.004MaRDI QIDQ6488717zbMATH OpenWikidataFDO

Authors Feihu Huang, Shangqian Gao

Publication date 17 October 2023

Published in Neural Networks (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/J.NEUNET.2022.06.004

zbMATH Keywords

composition optimizationdeep neural networksGrassmann manifoldprincipal component analysisRiemannian manifoldStiefel manifold

Mathematics Subject Classification ID

Differential geometry (53-XX) Statistics (62-XX) Artificial neural networks and deep learning (68T07)

Recommendations

• Riemannian Natural Gradient Methods
• Recent Advances in Stochastic Riemannian Optimization
• Riemannian stochastic variance-reduced cubic regularized Newton method for submanifold optimization
• Riemannian Stochastic Variance Reduced Gradient Algorithm with Retraction and Vector Transport
• Projection-free nonconvex stochastic optimization on Riemannian manifolds

Cites work

• A Broyden class of quasi-Newton methods for Riemannian optimization
• Accelerating Stochastic Composition Optimization
• Complete Dictionary Recovery Over the Sphere II: Recovery by Riemannian Trust-Region Method
• Distributionally robust optimization. A review on theory and applications
• Empirical Arithmetic Averaging Over the Compact Stiefel Manifold
• Gradient-based learning methods extended to smooth manifolds applied to automated clustering
• Lie-group-type neural system learning by manifold retractions
• Low-rank matrix completion by Riemannian optimization
• Manopt, a Matlab toolbox for optimization on manifolds
• Riemannian Stochastic Variance Reduced Gradient Algorithm with Retraction and Vector Transport
• Solving Stochastic Compositional Optimization is Nearly as Easy as Solving Stochastic Optimization

Cited in

(9)

• Stochastic completeness and gradient representations for Sub-Riemannian manifolds
• Recent Advances in Stochastic Riemannian Optimization
• Riemannian Stochastic Variance Reduced Gradient Algorithm with Retraction and Vector Transport
• Stochastic composition optimization of functions without Lipschitz continuous gradient
• Stochastic Zeroth-Order Riemannian Derivative Estimation and Optimization
• Riemannian Natural Gradient Methods
• Convergence of hyperbolic neural networks under Riemannian stochastic gradient descent
• Zeroth-order Riemannian averaging stochastic approximation algorithms
• Faster adaptive momentum-based federated methods for distributed composition optimization

This page was built for publication: Riemannian gradient methods for stochastic composition problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6488717)