On the computational efficiency of catalyst accelerated coordinate descent

DOI10.1007/978-3-030-77876-7_12zbMath1487.90526arXiv2103.06688OpenAlexW3176586272MaRDI QIDQ2117631

Dmitry Pasechnyuk, Vladislav Matyukhin

Publication date: 22 March 2022

Full work available at URL: https://arxiv.org/abs/2103.06688

zbMATH Keywords

Markov decision processescatalystSoftMaxaccelerated coordinate descent methodproximal accelerated method

Mathematics Subject Classification ID

Convex programming (90C25) Markov and semi-Markov decision processes (90C40)

Uses Software

• SciPy
• logsumexp-softmax-tests

Cites Work

• Unnamed Item
• Unnamed Item
• Smooth minimization of non-smooth functions
• Efficient numerical methods for entropy-linear programming problems
• Lectures on convex optimization
• An Accelerated Hybrid Proximal Extragradient Method for Convex Optimization and Its Implications to Second-Order Methods
• Efficiency of Coordinate Descent Methods on Huge-Scale Optimization Problems
• Efficiency of the Accelerated Coordinate Descent Method on Structured Optimization Problems
• Monotone Operators and the Proximal Point Algorithm
• Catalyst Acceleration for First-order Convex Optimization: from Theory to Practice
• Variance Reduced Value Iteration and Faster Algorithms for Solving Markov Decision Processes
• Contracting Proximal Methods for Smooth Convex Optimization

This page was built for publication: On the computational efficiency of catalyst accelerated coordinate descent