Making the Last Iterate of SGD Information Theoretically Optimal

DOI10.1137/19M128908XzbMath1466.90073arXiv1904.12443MaRDI QIDQ4987277

Prateek Jain, Praneeth Netrapalli, Dheeraj Nagaraj

Publication date: 3 May 2021

Published in: SIAM Journal on Optimization (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1904.12443

zbMATH Keywords

convex optimizationgradient descentstochastic gradient descentlast iterate guarantees

Mathematics Subject Classification ID

Convex programming (90C25) Stochastic programming (90C15)

Related Items (1)

Unnamed Item

Uses Software

• Pegasos

Cites Work

• Unnamed Item
• Unnamed Item
• Pegasos: primal estimated sub-gradient solver for SVM
• An optimal algorithm for stochastic strongly-convex optimization
• Large-Scale Machine Learning with Stochastic Gradient Descent
• On the Generalization Ability of On-Line Learning Algorithms
• Acceleration of Stochastic Approximation by Averaging
• Optimization Methods for Large-Scale Machine Learning
• A Stochastic Approximation Method

This page was built for publication: Making the Last Iterate of SGD Information Theoretically Optimal