DSCOVR: Randomized Primal-Dual Block Coordinate Algorithms for Asynchronous Distributed Optimization

zbMath1484.90081arXiv1710.05080MaRDI QIDQ4633058

Weizhu Chen, Qihang Lin, Lin Xiao, Adams Wei Yu

Publication date: 2 May 2019

Full work available at URL: https://arxiv.org/abs/1710.05080

zbMATH Keywords

variance reductionbig datadoubly stochastic coordinate optimizationlarge linear modelsoptimization variables

Mathematics Subject Classification ID

Generalized linear models (logistic models) (62J12) Convex programming (90C25) Learning and adaptive systems in artificial intelligence (68T05) Statistical aspects of big data and data science (62R07)

Related Items (8)

On lower iteration complexity bounds for the convex concave saddle point problems ⋮ Unnamed Item ⋮ Primal-Dual First-Order Methods for Affinely Constrained Multi-block Saddle Point Problems ⋮ Worst-case complexity of cyclic coordinate descent: \(O(n^2)\) gap with randomized version ⋮ An Optimal Algorithm for Decentralized Finite-Sum Optimization ⋮ DSCOVR ⋮ Unnamed Item ⋮ Unnamed Item

Uses Software

• LIBSVM
• MPI
• MLlib
• ARock
• HOGWILD
• AIDE
• DiSCO

This page was built for publication: DSCOVR: Randomized Primal-Dual Block Coordinate Algorithms for Asynchronous Distributed Optimization