Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
×
Recent work by Nesterov and Stich showed that momentum can be used to accelerate the rate of convergence for block Gauss-Seidel in the setting where a fixed partitioning of the coordinates is chosen ahead of time.
Jan 14, 2017
Empirically, we show that accelerated Gauss-Seidel with random coordinate sampling provides speedups for large scale machine learning tasks when compared to non ...
Sep 24, 2017 · Abstract. Recent work by Nesterov and Stich [15] showed that momentum can be used to accelerate the rate of convergence for block ...
Simple Goal: Fix an n×n (dense) positive definite matrix A and response vector b∈Rn. Our goal is to solve Ax=b. Caveat: we want n to be really big.
People also ask
Aug 6, 2017 · Motivated by this finding, we analyze the accelerated block Gauss-Seidel algorithm in the random coordinate sampling setting. Our analysis ...
Apr 8, 2020 · We present a novel greedy Gauss-Seidel method for solving large linear least squares problem. This method improves the greedy randomized ...
Breaking locality accelerates block Gauss-Seidel. S Tu, S Venkataraman, AC Wilson, A Gittens, MI Jordan, B Recht. International Conference on Machine Learning ...
2017. • “Breaking Locality Accelerates Block Gauss-Seidel.” Stephen Tu, Shivaram. Venkataraman, Ashia C. Wilson, Alex Gittens, Michael I. Jordan, and ...
May 22, 2021 · The analysis of the randomized Gauss–Seidel method and its extension shows that the larger the singular value of A is, the faster the error ...