Name
Abstract | ||
|---|---|---|
Various recent iterative optimization methods require to compute the proximity operator of a sum of functions. We address this problem by proposing a new distributed algorithm for a sum of non-necessarily smooth convex functions composed with arbitrary linear operators. In our approach, each function is associated with a node of a graph, which communicates with its neighbors. Our algorithm relies on a primal-dual splitting strategy that avoids to invert any linear operator, thus making it suitable for processing high-dimensional datasets. The proposed algorithm has a wide array of applications in signal/image processing and machine learning and its convergence is established. |
Year | DOI | Venue |
|---|---|---|
2015 | 10.1109/ACSSC.2015.7421156 | 2015 49th Asilomar Conference on Signals, Systems and Computers |
Keywords | Field | DocType |
proximity operators,iterative optimization methods,distributed algorithm,nonnecessarily-smooth convex functions,arbitrary linear operators,graph node,primal-dual splitting strategy,linear operator,high-dimensional dataset processing,signal processing,machine learning,convergence analysis,image processing | Mathematical optimization,Computer science,Image processing,Theoretical computer science,Distributed algorithm,Convex function,Operator (computer programming),Linear map,Convex optimization,Ellipsoid method,Linear matrix inequality | Conference |
Citations | PageRank | References |
0 | 0.34 | 16 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| F. Abboud | 1 | 3 | 0.73 |
| Emilie Chouzenoux | 2 | 202 | 26.37 |
| Jean-Christophe Pesquet | 3 | 560 | 46.10 |
| Jean-Hugues Chenot | 4 | 72 | 12.59 |
| Louis Laborelli | 5 | 80 | 11.78 |