Show simple item record

dc.contributor.authorMishchenko, Konstantin
dc.contributor.authorIutzeler, Franck
dc.contributor.authorMalick, Jérôme
dc.contributor.authorAmini, Massih Reza
dc.date.accessioned2020-12-29T12:35:23Z
dc.date.available2020-12-29T12:35:23Z
dc.date.issued2018-01-01
dc.identifier.isbn9781510867963
dc.identifier.urihttp://hdl.handle.net/10754/666756
dc.description.abstractDistributed learning aims at computing high- quality models by training over scattered data. This covers a diversity of scenarios, including computer clusters or mobile agents. One of the main challenges is then to deal with heterogeneous machines and unreliable communications. In this setting, we propose and analyze a flexible asynchronous optimization algorithm for solving nonsmooth learning problems. Unlike most existing methods, our algorithm is adjustable to various levels of communication costs, machines computational powers, and data distribution evenness. We prove that the algorithm converges linearly with a fixed learning rate that does not depend on communication delays nor on the number of machines. Although long delays in communication may slow down performance, no delay can break convergence.
dc.publisherInternational Machine Learning Society (IMLS)rasmussen@ptd.net
dc.relation.urlhttp://proceedings.mlr.press/v80/mishchenko18a.html
dc.rightsArchived with thanks to International Machine Learning Society (IMLS)rasmussen@ptd.net
dc.titleA delay-tolerant proximal-gradient algorithm for distributed learning
dc.typeConference Paper
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.conference.date2018-07-10 to 2018-07-15
dc.conference.name35th International Conference on Machine Learning, ICML 2018
dc.conference.locationStockholm, SWE
dc.eprint.versionPre-print
dc.contributor.institutionUniv. Grenoble Alpes, France
dc.contributor.institutionCNRS and LJK, France
dc.identifier.volume8
dc.identifier.pages5774-5788
kaust.personMishchenko, Konstantin
dc.identifier.eid2-s2.0-85057234444


This item appears in the following Collection(s)

Show simple item record