Matches in SemOpenAlex for { <https://semopenalex.org/work/W2948215117> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W2948215117 abstract "We consider the fully decentralized machine learning scenario where many users with personal datasets collaborate to learn models through local peer-to-peer exchanges, without a central coordinator. We propose to train personalized models that leverage a collaboration graph describing the relationships between user personal tasks, which we learn jointly with the models. Our fully decentralized optimization procedure alternates between training nonlinear models given the graph in a greedy boosting manner, and updating the collaboration graph (with controlled sparsity) given the models. Throughout the process, users exchange messages only with a small number of peers (their direct neighbors when updating the models, and a few random users when updating the graph), ensuring that the procedure naturally scales with the number of users. Overall, our approach is communication-efficient and avoids exchanging personal data. We provide an extensive analysis of the convergence rate, memory and communication complexity of our approach, and demonstrate its benefits compared to competing techniques on synthetic and real datasets." @default.
- W2948215117 created "2019-06-14" @default.
- W2948215117 creator A5003859975 @default.
- W2948215117 creator A5014504793 @default.
- W2948215117 creator A5047374978 @default.
- W2948215117 date "2020-08-26" @default.
- W2948215117 modified "2023-09-29" @default.
- W2948215117 title "Fully Decentralized Joint Learning of Personalized Models and Collaboration Graphs" @default.
- W2948215117 cites W134960717 @default.
- W2948215117 cites W1540764732 @default.
- W2948215117 cites W1541700983 @default.
- W2948215117 cites W1678356000 @default.
- W2948215117 cites W1971125461 @default.
- W2948215117 cites W1988790447 @default.
- W2948215117 cites W2000769684 @default.
- W2948215117 cites W2046387702 @default.
- W2948215117 cites W2050968963 @default.
- W2948215117 cites W2072265605 @default.
- W2948215117 cites W2109706083 @default.
- W2948215117 cites W2117686388 @default.
- W2948215117 cites W2117905067 @default.
- W2948215117 cites W2125607229 @default.
- W2948215117 cites W2126147068 @default.
- W2948215117 cites W2136885855 @default.
- W2948215117 cites W2157017541 @default.
- W2948215117 cites W2187705271 @default.
- W2948215117 cites W2212091021 @default.
- W2948215117 cites W2236749180 @default.
- W2948215117 cites W2295943972 @default.
- W2948215117 cites W2407738787 @default.
- W2948215117 cites W2535838896 @default.
- W2948215117 cites W2539283139 @default.
- W2948215117 cites W2541884796 @default.
- W2948215117 cites W2742079690 @default.
- W2948215117 cites W2781299994 @default.
- W2948215117 cites W2787878670 @default.
- W2948215117 cites W2789243870 @default.
- W2948215117 cites W2800334336 @default.
- W2948215117 cites W2803972986 @default.
- W2948215117 cites W2911978475 @default.
- W2948215117 cites W2959406683 @default.
- W2948215117 cites W2962863496 @default.
- W2948215117 cites W2963191611 @default.
- W2948215117 cites W2963228337 @default.
- W2948215117 cites W2963300197 @default.
- W2948215117 cites W2963712608 @default.
- W2948215117 cites W2963855915 @default.
- W2948215117 cites W2964012239 @default.
- W2948215117 cites W2964262905 @default.
- W2948215117 cites W2995022099 @default.
- W2948215117 cites W3099453040 @default.
- W2948215117 cites W3101665129 @default.
- W2948215117 hasPublicationYear "2020" @default.
- W2948215117 type Work @default.
- W2948215117 sameAs 2948215117 @default.
- W2948215117 citedByCount "2" @default.
- W2948215117 countsByYear W29482151172019 @default.
- W2948215117 countsByYear W29482151172020 @default.
- W2948215117 crossrefType "proceedings-article" @default.
- W2948215117 hasAuthorship W2948215117A5003859975 @default.
- W2948215117 hasAuthorship W2948215117A5014504793 @default.
- W2948215117 hasAuthorship W2948215117A5047374978 @default.
- W2948215117 hasBestOaLocation W29482151171 @default.
- W2948215117 hasConcept C127413603 @default.
- W2948215117 hasConcept C154945302 @default.
- W2948215117 hasConcept C170154142 @default.
- W2948215117 hasConcept C18555067 @default.
- W2948215117 hasConcept C2522767166 @default.
- W2948215117 hasConcept C41008148 @default.
- W2948215117 hasConcept C80444323 @default.
- W2948215117 hasConceptScore W2948215117C127413603 @default.
- W2948215117 hasConceptScore W2948215117C154945302 @default.
- W2948215117 hasConceptScore W2948215117C170154142 @default.
- W2948215117 hasConceptScore W2948215117C18555067 @default.
- W2948215117 hasConceptScore W2948215117C2522767166 @default.
- W2948215117 hasConceptScore W2948215117C41008148 @default.
- W2948215117 hasConceptScore W2948215117C80444323 @default.
- W2948215117 hasLocation W29482151171 @default.
- W2948215117 hasLocation W29482151172 @default.
- W2948215117 hasLocation W29482151173 @default.
- W2948215117 hasLocation W29482151174 @default.
- W2948215117 hasOpenAccess W2948215117 @default.
- W2948215117 hasPrimaryLocation W29482151171 @default.
- W2948215117 hasRelatedWork W2027246577 @default.
- W2948215117 hasRelatedWork W2030530201 @default.
- W2948215117 hasRelatedWork W2351267244 @default.
- W2948215117 hasRelatedWork W2790424990 @default.
- W2948215117 hasRelatedWork W2912695858 @default.
- W2948215117 hasRelatedWork W3011555937 @default.
- W2948215117 hasRelatedWork W3096054746 @default.
- W2948215117 hasRelatedWork W3107474891 @default.
- W2948215117 hasRelatedWork W3199858452 @default.
- W2948215117 hasRelatedWork W4312221968 @default.
- W2948215117 isParatext "false" @default.
- W2948215117 isRetracted "false" @default.
- W2948215117 magId "2948215117" @default.
- W2948215117 workType "article" @default.