dc.contributor.author | Gratton, Cristiano | |
dc.contributor.author | Dasanadoddi Venkategowda, Naveen Kumar | |
dc.contributor.author | Arablouei, Reza | |
dc.contributor.author | Werner, Stefan | |
dc.date.accessioned | 2021-02-23T09:44:25Z | |
dc.date.available | 2021-02-23T09:44:25Z | |
dc.date.created | 2021-01-11T16:34:56Z | |
dc.date.issued | 2020 | |
dc.identifier.isbn | 978-9-0827-9705-3 | |
dc.identifier.uri | https://hdl.handle.net/11250/2729696 | |
dc.description.abstract | We develop a new distributed algorithm to solve a learning problem with non-smooth objective functions when data are distributed over a multi-agent network. We employ a zeroth-order method to minimize the associated augmented Lagrangian in the primal domain using the alternating direction method of multipliers (ADMM) to develop the proposed algorithm, named distributed zeroth-order based ADMM (D-ZOA). Unlike most existing algorithms for non-smooth optimization, which rely on calculating subgradients or proximal operators, D-ZOA only requires function values to approximate gradients of the objective function. Convergence of D-ZOA to the centralized solution is confirmed via theoretical analysis and simulation results. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | IEEE | en_US |
dc.relation.ispartof | 28th European Signal Processing Conference (EUSIPCO 2020) | |
dc.title | Distributed Learning with Non-Smooth Objective Functions | en_US |
dc.type | Chapter | en_US |
dc.description.version | acceptedVersion | en_US |
dc.source.pagenumber | 2180-2184 | en_US |
dc.identifier.doi | https://doi.org/10.23919/Eusipco47968.2020.9287441 | |
dc.identifier.cristin | 1869275 | |
dc.description.localcode | © 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. | en_US |
cristin.ispublished | true | |
cristin.fulltext | postprint | |