FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3090221383> ?p ?o ?g. }

• W3090221383 abstract "We design a low complexity decentralized learning algorithm to train a recently proposed large neural network in distributed processing nodes (workers). We assume the communication network between the workers is synchronized and can be modeled as a doubly-stochastic mixing matrix without having any master node. In our setup, the training data is distributed among the workers but is not shared in the training process due to privacy and security concerns. Using alternating-direction-method-of-multipliers (ADMM) along with a layerwise convex optimization approach, we propose a decentralized learning algorithm which enjoys low computational complexity and communication cost among the workers. We show that it is possible to achieve equivalent learning performance as if the data is available in a single place. Finally, we experimentally illustrate the time complexity and convergence behavior of the algorithm." @default.
• W3090221383 created "2020-10-08" @default.
• W3090221383 creator A5031351754 @default.
• W3090221383 creator A5041348422 @default.
• W3090221383 creator A5043099254 @default.
• W3090221383 creator A5048010798 @default.
• W3090221383 date "2020-09-29" @default.
• W3090221383 modified "2023-10-01" @default.
• W3090221383 title "A Low Complexity Decentralized Neural Net with Centralized Equivalence using Layer-wise Learning" @default.
• W3090221383 cites W1181965167 @default.
• W3090221383 cites W1439842103 @default.
• W3090221383 cites W1582862366 @default.
• W3090221383 cites W1988224834 @default.
• W3090221383 cites W2052842899 @default.
• W3090221383 cites W2084251795 @default.
• W3090221383 cites W2084716923 @default.
• W3090221383 cites W2101517602 @default.
• W3090221383 cites W2107438106 @default.
• W3090221383 cites W2111072639 @default.
• W3090221383 cites W2112796928 @default.
• W3090221383 cites W2134557905 @default.
• W3090221383 cites W2138383519 @default.
• W3090221383 cites W2164278908 @default.
• W3090221383 cites W2168231600 @default.
• W3090221383 cites W2173213060 @default.
• W3090221383 cites W2301541953 @default.
• W3090221383 cites W2346438296 @default.
• W3090221383 cites W2413539728 @default.
• W3090221383 cites W2553581924 @default.
• W3090221383 cites W2562947506 @default.
• W3090221383 cites W2766674581 @default.
• W3090221383 cites W2786070938 @default.
• W3090221383 cites W2787017828 @default.
• W3090221383 cites W2890010271 @default.
• W3090221383 cites W2890924858 @default.
• W3090221383 cites W2913664433 @default.
• W3090221383 cites W2917189468 @default.
• W3090221383 cites W2944894845 @default.
• W3090221383 cites W2962741697 @default.
• W3090221383 cites W2963209930 @default.
• W3090221383 cites W2963773265 @default.
• W3090221383 cites W2970404488 @default.
• W3090221383 cites W2981174158 @default.
• W3090221383 cites W3101665129 @default.
• W3090221383 cites W3120740533 @default.
• W3090221383 cites W604355987 @default.
• W3090221383 doi "https://doi.org/10.48550/arxiv.2009.13982" @default.
• W3090221383 hasPublicationYear "2020" @default.
• W3090221383 type Work @default.
• W3090221383 sameAs 3090221383 @default.
• W3090221383 citedByCount "1" @default.
• W3090221383 countsByYear W30902213832020 @default.
• W3090221383 crossrefType "posted-content" @default.
• W3090221383 hasAuthorship W3090221383A5031351754 @default.
• W3090221383 hasAuthorship W3090221383A5041348422 @default.
• W3090221383 hasAuthorship W3090221383A5043099254 @default.
• W3090221383 hasAuthorship W3090221383A5048010798 @default.
• W3090221383 hasBestOaLocation W30902213831 @default.
• W3090221383 hasConcept C111919701 @default.
• W3090221383 hasConcept C11413529 @default.
• W3090221383 hasConcept C118615104 @default.
• W3090221383 hasConcept C120314980 @default.
• W3090221383 hasConcept C126255220 @default.
• W3090221383 hasConcept C127413603 @default.
• W3090221383 hasConcept C154945302 @default.
• W3090221383 hasConcept C162324750 @default.
• W3090221383 hasConcept C178790620 @default.
• W3090221383 hasConcept C179145077 @default.
• W3090221383 hasConcept C179799912 @default.
• W3090221383 hasConcept C185592680 @default.
• W3090221383 hasConcept C2777303404 @default.
• W3090221383 hasConcept C2779227376 @default.
• W3090221383 hasConcept C2780069185 @default.
• W3090221383 hasConcept C33923547 @default.
• W3090221383 hasConcept C41008148 @default.
• W3090221383 hasConcept C50522688 @default.
• W3090221383 hasConcept C50644808 @default.
• W3090221383 hasConcept C62611344 @default.
• W3090221383 hasConcept C66938386 @default.
• W3090221383 hasConcept C80444323 @default.
• W3090221383 hasConcept C98045186 @default.
• W3090221383 hasConceptScore W3090221383C111919701 @default.
• W3090221383 hasConceptScore W3090221383C11413529 @default.
• W3090221383 hasConceptScore W3090221383C118615104 @default.
• W3090221383 hasConceptScore W3090221383C120314980 @default.
• W3090221383 hasConceptScore W3090221383C126255220 @default.
• W3090221383 hasConceptScore W3090221383C127413603 @default.
• W3090221383 hasConceptScore W3090221383C154945302 @default.
• W3090221383 hasConceptScore W3090221383C162324750 @default.
• W3090221383 hasConceptScore W3090221383C178790620 @default.
• W3090221383 hasConceptScore W3090221383C179145077 @default.
• W3090221383 hasConceptScore W3090221383C179799912 @default.
• W3090221383 hasConceptScore W3090221383C185592680 @default.
• W3090221383 hasConceptScore W3090221383C2777303404 @default.
• W3090221383 hasConceptScore W3090221383C2779227376 @default.
• W3090221383 hasConceptScore W3090221383C2780069185 @default.
• W3090221383 hasConceptScore W3090221383C33923547 @default.
• W3090221383 hasConceptScore W3090221383C41008148 @default.
• W3090221383 hasConceptScore W3090221383C50522688 @default.
• W3090221383 hasConceptScore W3090221383C50644808 @default.

• next