FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2148844117> ?p ?o ?g. }

• W2148844117 abstract "Scalable machine learning over big data stored on a cluster of commodity machines with significant communication costs has become important in recent years. In this paper we give a novel approach to the distributed training of linear classifiers (involving smooth losses and L2 regularization) that is designed to reduce communication costs. At each iteration, the nodes minimize approximate objective functions; then the resulting minimizers are combined to form a descent direction to move. Our approach gives a lot of freedom in the formation of the approximate objective function as well as in the choice of methods to solve them. The method is shown to have O(log(1/ǫ)) time convergence. The method can be viewed as an iterative parameter mixing method. A special instantiation yields a parallel stochastic gradient descent method with strong convergence. When communication times between nodes are large, our method is much faster than the SQM method [7], which computes function and gradient values in a distributed fashion." @default.
• W2148844117 created "2016-06-24" @default.
• W2148844117 creator A5004541836 @default.
• W2148844117 creator A5019206666 @default.
• W2148844117 creator A5068664439 @default.
• W2148844117 creator A5090165776 @default.
• W2148844117 date "2013-10-31" @default.
• W2148844117 modified "2023-09-26" @default.
• W2148844117 title "A Functional Approximation Based Distributed Learning Algorithm." @default.
• W2148844117 cites W114517082 @default.
• W2148844117 cites W1549918636 @default.
• W2148844117 cites W1562070494 @default.
• W2148844117 cites W1603765807 @default.
• W2148844117 cites W1606153355 @default.
• W2148844117 cites W1988849934 @default.
• W2148844117 cites W2012960907 @default.
• W2148844117 cites W2035815925 @default.
• W2148844117 cites W2051434435 @default.
• W2148844117 cites W2106833843 @default.
• W2148844117 cites W2107438106 @default.
• W2148844117 cites W2109722477 @default.
• W2148844117 cites W2117669906 @default.
• W2148844117 cites W2120350480 @default.
• W2148844117 cites W2128529555 @default.
• W2148844117 cites W2129191766 @default.
• W2148844117 cites W2138302120 @default.
• W2148844117 cites W2144839971 @default.
• W2148844117 cites W2157462866 @default.
• W2148844117 cites W2164278908 @default.
• W2148844117 cites W2165966284 @default.
• W2148844117 cites W2166706236 @default.
• W2148844117 cites W2296319761 @default.
• W2148844117 cites W2311729496 @default.
• W2148844117 cites W2951977692 @default.
• W2148844117 hasPublicationYear "2013" @default.
• W2148844117 type Work @default.
• W2148844117 sameAs 2148844117 @default.
• W2148844117 citedByCount "6" @default.
• W2148844117 countsByYear W21488441172013 @default.
• W2148844117 countsByYear W21488441172015 @default.
• W2148844117 countsByYear W21488441172016 @default.
• W2148844117 countsByYear W21488441172017 @default.
• W2148844117 countsByYear W21488441172019 @default.
• W2148844117 crossrefType "posted-content" @default.
• W2148844117 hasAuthorship W2148844117A5004541836 @default.
• W2148844117 hasAuthorship W2148844117A5019206666 @default.
• W2148844117 hasAuthorship W2148844117A5068664439 @default.
• W2148844117 hasAuthorship W2148844117A5090165776 @default.
• W2148844117 hasConcept C11413529 @default.
• W2148844117 hasConcept C126255220 @default.
• W2148844117 hasConcept C14036430 @default.
• W2148844117 hasConcept C153258448 @default.
• W2148844117 hasConcept C154945302 @default.
• W2148844117 hasConcept C15744967 @default.
• W2148844117 hasConcept C162324750 @default.
• W2148844117 hasConcept C19417346 @default.
• W2148844117 hasConcept C206688291 @default.
• W2148844117 hasConcept C2776135515 @default.
• W2148844117 hasConcept C2777303404 @default.
• W2148844117 hasConcept C2779582901 @default.
• W2148844117 hasConcept C33923547 @default.
• W2148844117 hasConcept C41008148 @default.
• W2148844117 hasConcept C48044578 @default.
• W2148844117 hasConcept C50522688 @default.
• W2148844117 hasConcept C50644808 @default.
• W2148844117 hasConcept C77088390 @default.
• W2148844117 hasConcept C78458016 @default.
• W2148844117 hasConcept C86803240 @default.
• W2148844117 hasConceptScore W2148844117C11413529 @default.
• W2148844117 hasConceptScore W2148844117C126255220 @default.
• W2148844117 hasConceptScore W2148844117C14036430 @default.
• W2148844117 hasConceptScore W2148844117C153258448 @default.
• W2148844117 hasConceptScore W2148844117C154945302 @default.
• W2148844117 hasConceptScore W2148844117C15744967 @default.
• W2148844117 hasConceptScore W2148844117C162324750 @default.
• W2148844117 hasConceptScore W2148844117C19417346 @default.
• W2148844117 hasConceptScore W2148844117C206688291 @default.
• W2148844117 hasConceptScore W2148844117C2776135515 @default.
• W2148844117 hasConceptScore W2148844117C2777303404 @default.
• W2148844117 hasConceptScore W2148844117C2779582901 @default.
• W2148844117 hasConceptScore W2148844117C33923547 @default.
• W2148844117 hasConceptScore W2148844117C41008148 @default.
• W2148844117 hasConceptScore W2148844117C48044578 @default.
• W2148844117 hasConceptScore W2148844117C50522688 @default.
• W2148844117 hasConceptScore W2148844117C50644808 @default.
• W2148844117 hasConceptScore W2148844117C77088390 @default.
• W2148844117 hasConceptScore W2148844117C78458016 @default.
• W2148844117 hasConceptScore W2148844117C86803240 @default.
• W2148844117 hasLocation W21488441171 @default.
• W2148844117 hasOpenAccess W2148844117 @default.
• W2148844117 hasPrimaryLocation W21488441171 @default.
• W2148844117 hasRelatedWork W2106955134 @default.
• W2148844117 hasRelatedWork W2164278908 @default.
• W2148844117 hasRelatedWork W2226884960 @default.
• W2148844117 hasRelatedWork W2396373224 @default.
• W2148844117 hasRelatedWork W2400034325 @default.
• W2148844117 hasRelatedWork W2469560512 @default.
• W2148844117 hasRelatedWork W2520915071 @default.
• W2148844117 hasRelatedWork W2556206132 @default.
• W2148844117 hasRelatedWork W2910169538 @default.

• next