On the linear convergence of the stochastic gradient method with constant step-size

DOI10.1007/s11590-018-1331-1zbMath1430.90440arXiv1712.01906MaRDI QIDQ2311205

Băng Công Vũ, Volkan Cevher

Publication date: 10 July 2019

Published in: Optimization Letters (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1712.01906

zbMATH Keywords

linear convergence; stochastic gradient; strong growth condition

Mathematics Subject Classification ID

90C25: Convex programming

90C15: Stochastic programming

Related Items

Accelerated stochastic variance reduction for a class of convex optimization problems

Cites Work

• Unnamed Item
• Stochastic forward-backward splitting for monotone inclusions
• A randomized Kaczmarz algorithm with exponential convergence
• Incremental gradient algorithms with stepsizes bounded away from zero
• A globally convergent incremental Newton method
• Restricted strong convexity and its applications to convergence analysis of gradient-type methods in convex optimization
• Stochastic Approximations and Perturbations in Forward-Backward Splitting for Monotone Operators
• Robust Stochastic Approximation Approach to Stochastic Programming
• An Incremental Gradient(-Projection) Method with Momentum Term and Adaptive Stepsize Rule
• A Proximal Stochastic Gradient Method with Progressive Variance Reduction
• Convex analysis and monotone operator theory in Hilbert spaces
• Stochastic gradient descent, weighted sampling, and the randomized Kaczmarz algorithm