An accelerated communication-efficient primal-dual optimization framework for structured machine learning

DOI10.1080/10556788.2019.1650361zbMath1464.90059arXiv1711.05305OpenAlexW2966396637WikidataQ127402834 ScholiaQ127402834MaRDI QIDQ5859008

Martin Takáč, Chenxin Ma, Martin Jaggi, Nathan Srebro, Frank E. Curtis

Publication date: 15 April 2021

Published in: Optimization Methods and Software (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1711.05305

zbMATH Keywords

nonsmooth optimizationnonlinear optimizationmachine learningdistributed optimizationaccelerated methods

Mathematics Subject Classification ID

Convex programming (90C25)

Related Items

Communication-efficient distributed multi-task learning with matrix sparsity regularization, Distributed Learning with Sparse Communications by Identification, Fast and safe: accelerated gradient methods with optimality certificates and underestimate sequences

Uses Software

• CoCoA
• L-BFGS-B
• L-BFGS
• LBFGS-B
• AdaGrad
• HOGWILD

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
• Gradient methods for minimizing composite functions
• Communication-efficient distributed optimization of self-concordant empirical loss
• Asynchronous Stochastic Coordinate Descent: Parallelism and Convergence Properties
• Accelerated, Parallel, and Proximal Coordinate Descent
• Algorithm 778: L-BFGS-B
• Distributed optimization with arbitrary local solvers
• Stochastic Dual Coordinate Ascent Methods for Regularized Loss Minimization