DocumentCode :
180222
Title :
Flexible parallel algorithms for big data optimization
Author :
Facchinei, Francisco ; Sagratella, Simone ; Scutari, Gesualdo
Author_Institution :
Dept. of Comput., Control, & Manage. Eng., Univ. of Rome La Sapienza, Rome, Italy
fYear :
2014
fDate :
4-9 May 2014
Firstpage :
7208
Lastpage :
7212
Abstract :
We propose a decomposition framework for the parallel optimization of the sum of a differentiable function and a (block) separable nonsmooth, convex one. The latter term is typically used to enforce structure in the solution as, for example, in LASSO problems. Our framework is very flexible and includes both fully parallel Jacobi schemes and Gauss-Seidel (Southwell-type) ones, as well as virtually all possibilities in between (e.g., gradient- or Newton-type methods) with only a subset of variables updated at each iteration. Our theoretical convergence results improve on existing ones, and numerical results show that the new method compares favorably to existing algorithms.
Keywords :
Big Data; iterative methods; minimisation; parallel algorithms; Big Data optimization; Gauss-Seidel schemes; LASSO problems; decomposition framework; flexible parallel algorithms; fully parallel Jacobi schemes; parallel optimization; Approximation algorithms; Approximation methods; Convergence; Jacobian matrices; Minimization; Optimization; Parallel algorithms; Jacobi method; LASSO; Parallel optimization; Sparse solution;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2014 IEEE International Conference on
Conference_Location :
Florence
Type :
conf
DOI :
10.1109/ICASSP.2014.6854999
Filename :
6854999
Link To Document :
بازگشت