Matches in SemOpenAlex for { <https://semopenalex.org/work/W2790352117> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W2790352117 abstract "Author(s): Pan, Xinghao | Advisor(s): Jordan, Michael I | Abstract: Many machine learning algorithms iteratively process datapoints and transform global model parameters. It has become increasingly impractical to serially execute such iterative algorithms as processor speeds fail to catch up to the growth in dataset sizes.To address these problems, the machine learning community has turned to two parallelization strategies: bulk synchronous parallel (BSP), and coordination-free. BSP algorithms partition computational work among workers, with occasional synchronization at global barriers, but has only been applied to ‘embarrassingly parallel’ problems where work is trivially factorizable. Coordination-free algorithms simply allow concurrent processors to execute in parallel, interleaving transformations and possibly introducing inconsistencies. Theoretical analysis is then required to prove that the coordination-free algorithm produces a reasonable approximation to the desired outcome, under assumptions on the problem and system.In this dissertation, we propose and explore a third approach by applying concurrency control to manage parallel transformations in machine learning algorithms. We identify points of possible interference between parallel iterations by examining the semantics of the serial algorithm. Coordination is then introduced to either avoid or resolve such conflicts, whereas non-conflicting transformations are allowed to execute concurrently. Our parallel algorithms are thus engineered to produce the same exact output as the serial machine learning algorithm, preserving the serial algorithm’s theoretical guarantees of correctness while maximizing concurrency.We demonstrate the feasibility of our approach to parallelizing a variety of machine learning algorithms, including nonparametric unsupervised learning, graph clustering, discrete optimization, and sparse convex optimization. We theoretically prove and empirically verify that our parallel algorithms produce equivalent output to their serial counterparts. We also theoretically analyze the expected concurrency of our parallel algorithms, and empirically demonstrate their scalability." @default.
- W2790352117 created "2018-03-29" @default.
- W2790352117 creator A5039564077 @default.
- W2790352117 date "2017-01-01" @default.
- W2790352117 modified "2023-09-26" @default.
- W2790352117 title "Parallel Machine Learning Using Concurrency Control" @default.
- W2790352117 cites W2094156322 @default.
- W2790352117 cites W2166706236 @default.
- W2790352117 cites W2402144811 @default.
- W2790352117 cites W2949198759 @default.
- W2790352117 cites W2951781666 @default.
- W2790352117 hasPublicationYear "2017" @default.
- W2790352117 type Work @default.
- W2790352117 sameAs 2790352117 @default.
- W2790352117 citedByCount "1" @default.
- W2790352117 countsByYear W27903521172018 @default.
- W2790352117 crossrefType "journal-article" @default.
- W2790352117 hasAuthorship W2790352117A5039564077 @default.
- W2790352117 hasConcept C11413529 @default.
- W2790352117 hasConcept C119857082 @default.
- W2790352117 hasConcept C120314980 @default.
- W2790352117 hasConcept C120373497 @default.
- W2790352117 hasConcept C126909462 @default.
- W2790352117 hasConcept C154945302 @default.
- W2790352117 hasConcept C156891508 @default.
- W2790352117 hasConcept C173608175 @default.
- W2790352117 hasConcept C193702766 @default.
- W2790352117 hasConcept C41008148 @default.
- W2790352117 hasConcept C73555534 @default.
- W2790352117 hasConcept C80444323 @default.
- W2790352117 hasConceptScore W2790352117C11413529 @default.
- W2790352117 hasConceptScore W2790352117C119857082 @default.
- W2790352117 hasConceptScore W2790352117C120314980 @default.
- W2790352117 hasConceptScore W2790352117C120373497 @default.
- W2790352117 hasConceptScore W2790352117C126909462 @default.
- W2790352117 hasConceptScore W2790352117C154945302 @default.
- W2790352117 hasConceptScore W2790352117C156891508 @default.
- W2790352117 hasConceptScore W2790352117C173608175 @default.
- W2790352117 hasConceptScore W2790352117C193702766 @default.
- W2790352117 hasConceptScore W2790352117C41008148 @default.
- W2790352117 hasConceptScore W2790352117C73555534 @default.
- W2790352117 hasConceptScore W2790352117C80444323 @default.
- W2790352117 hasLocation W27903521171 @default.
- W2790352117 hasOpenAccess W2790352117 @default.
- W2790352117 hasPrimaryLocation W27903521171 @default.
- W2790352117 hasRelatedWork W1545030571 @default.
- W2790352117 hasRelatedWork W1607973859 @default.
- W2790352117 hasRelatedWork W2062590409 @default.
- W2790352117 hasRelatedWork W207818447 @default.
- W2790352117 hasRelatedWork W2147799885 @default.
- W2790352117 hasRelatedWork W2273481398 @default.
- W2790352117 hasRelatedWork W2587322366 @default.
- W2790352117 hasRelatedWork W2594063632 @default.
- W2790352117 hasRelatedWork W2769788175 @default.
- W2790352117 hasRelatedWork W2884886796 @default.
- W2790352117 hasRelatedWork W2914774799 @default.
- W2790352117 hasRelatedWork W2954005332 @default.
- W2790352117 hasRelatedWork W2963002787 @default.
- W2790352117 hasRelatedWork W2985253298 @default.
- W2790352117 hasRelatedWork W3126635575 @default.
- W2790352117 hasRelatedWork W3149606206 @default.
- W2790352117 hasRelatedWork W3154989773 @default.
- W2790352117 hasRelatedWork W71000211 @default.
- W2790352117 hasRelatedWork W118146258 @default.
- W2790352117 hasRelatedWork W1562636605 @default.
- W2790352117 isParatext "false" @default.
- W2790352117 isRetracted "false" @default.
- W2790352117 magId "2790352117" @default.
- W2790352117 workType "article" @default.