Matches in SemOpenAlex for { <https://semopenalex.org/work/W2779248810> ?p ?o ?g. }
- W2779248810 endingPage "1475" @default.
- W2779248810 startingPage "1457" @default.
- W2779248810 abstract "We consider the problem of learning a high-dimensional but low-rank matrix from a large-scale dataset distributed over several machines, where low-rankness is enforced by a convex trace norm constraint. We propose DFW-Trace, a distributed Frank-Wolfe algorithm which leverages the low-rank structure of its updates to achieve efficiency in time, memory and communication usage. The step at the heart of DFW-Trace is solved approximately using a distributed version of the power method. We provide a theoretical analysis of the convergence of DFW-Trace, showing that we can ensure sublinear convergence in expectation to an optimal solution with few power iterations per epoch. We implement DFW-Trace in the Apache Spark distributed programming framework and validate the usefulness of our approach on synthetic and real data, including the ImageNet dataset with high-dimensional features extracted from a deep neural network." @default.
- W2779248810 created "2018-01-05" @default.
- W2779248810 creator A5013781210 @default.
- W2779248810 creator A5014504793 @default.
- W2779248810 creator A5086752907 @default.
- W2779248810 date "2018-05-10" @default.
- W2779248810 modified "2023-10-17" @default.
- W2779248810 title "A distributed Frank–Wolfe framework for learning low-rank matrices with the trace norm" @default.
- W2779248810 cites W1529624360 @default.
- W2779248810 cites W1775587472 @default.
- W2779248810 cites W1967073510 @default.
- W2779248810 cites W1999913571 @default.
- W2779248810 cites W2003146967 @default.
- W2779248810 cites W2011359124 @default.
- W2779248810 cites W2040466831 @default.
- W2779248810 cites W2045704273 @default.
- W2779248810 cites W2054141820 @default.
- W2779248810 cites W2065180801 @default.
- W2779248810 cites W2076095618 @default.
- W2779248810 cites W2102019642 @default.
- W2779248810 cites W2103972604 @default.
- W2779248810 cites W2108598243 @default.
- W2779248810 cites W2109240917 @default.
- W2779248810 cites W2109521352 @default.
- W2779248810 cites W2109706083 @default.
- W2779248810 cites W2110531331 @default.
- W2779248810 cites W2117539524 @default.
- W2779248810 cites W2134332047 @default.
- W2779248810 cites W2136885855 @default.
- W2779248810 cites W2151148091 @default.
- W2779248810 cites W2151211319 @default.
- W2779248810 cites W2173213060 @default.
- W2779248810 cites W2188195356 @default.
- W2779248810 cites W2194775991 @default.
- W2779248810 cites W2200752277 @default.
- W2779248810 cites W2604884598 @default.
- W2779248810 cites W2611328865 @default.
- W2779248810 cites W2772486643 @default.
- W2779248810 cites W2913340405 @default.
- W2779248810 cites W2962710568 @default.
- W2779248810 cites W2962769133 @default.
- W2779248810 cites W3104501706 @default.
- W2779248810 cites W3104871901 @default.
- W2779248810 doi "https://doi.org/10.1007/s10994-018-5713-5" @default.
- W2779248810 hasPublicationYear "2018" @default.
- W2779248810 type Work @default.
- W2779248810 sameAs 2779248810 @default.
- W2779248810 citedByCount "8" @default.
- W2779248810 countsByYear W27792488102018 @default.
- W2779248810 countsByYear W27792488102021 @default.
- W2779248810 countsByYear W27792488102022 @default.
- W2779248810 crossrefType "journal-article" @default.
- W2779248810 hasAuthorship W2779248810A5013781210 @default.
- W2779248810 hasAuthorship W2779248810A5014504793 @default.
- W2779248810 hasAuthorship W2779248810A5086752907 @default.
- W2779248810 hasBestOaLocation W27792488101 @default.
- W2779248810 hasConcept C11413529 @default.
- W2779248810 hasConcept C117160843 @default.
- W2779248810 hasConcept C118615104 @default.
- W2779248810 hasConcept C121332964 @default.
- W2779248810 hasConcept C138885662 @default.
- W2779248810 hasConcept C162324750 @default.
- W2779248810 hasConcept C163716315 @default.
- W2779248810 hasConcept C2777303404 @default.
- W2779248810 hasConcept C2778459887 @default.
- W2779248810 hasConcept C33923547 @default.
- W2779248810 hasConcept C41008148 @default.
- W2779248810 hasConcept C41895202 @default.
- W2779248810 hasConcept C50522688 @default.
- W2779248810 hasConcept C62520636 @default.
- W2779248810 hasConcept C75291252 @default.
- W2779248810 hasConcept C80444323 @default.
- W2779248810 hasConceptScore W2779248810C11413529 @default.
- W2779248810 hasConceptScore W2779248810C117160843 @default.
- W2779248810 hasConceptScore W2779248810C118615104 @default.
- W2779248810 hasConceptScore W2779248810C121332964 @default.
- W2779248810 hasConceptScore W2779248810C138885662 @default.
- W2779248810 hasConceptScore W2779248810C162324750 @default.
- W2779248810 hasConceptScore W2779248810C163716315 @default.
- W2779248810 hasConceptScore W2779248810C2777303404 @default.
- W2779248810 hasConceptScore W2779248810C2778459887 @default.
- W2779248810 hasConceptScore W2779248810C33923547 @default.
- W2779248810 hasConceptScore W2779248810C41008148 @default.
- W2779248810 hasConceptScore W2779248810C41895202 @default.
- W2779248810 hasConceptScore W2779248810C50522688 @default.
- W2779248810 hasConceptScore W2779248810C62520636 @default.
- W2779248810 hasConceptScore W2779248810C75291252 @default.
- W2779248810 hasConceptScore W2779248810C80444323 @default.
- W2779248810 hasFunder F4320320883 @default.
- W2779248810 hasIssue "8-10" @default.
- W2779248810 hasLocation W27792488101 @default.
- W2779248810 hasLocation W277924881010 @default.
- W2779248810 hasLocation W277924881011 @default.
- W2779248810 hasLocation W277924881012 @default.
- W2779248810 hasLocation W27792488102 @default.
- W2779248810 hasLocation W27792488103 @default.
- W2779248810 hasLocation W27792488104 @default.
- W2779248810 hasLocation W27792488105 @default.