Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366196580> ?p ?o ?g. }
- W4366196580 abstract "Federated Recommendation can mitigate the systematical privacy risks of traditional recommendation since it allows the model training and online inferring without centralized user data collection. Most existing works assume that all user devices are available and adequate to participate in the Federated Learning. However, in practice, the complex recommendation models designed for accurate prediction and massive item data cause a high computation and communication cost to the resource-constrained user device, resulting in poor performance or training failure. Therefore, how to effectively compress the computation and communication overhead to achieve efficient federated recommendations across ubiquitous mobile devices remains a significant challenge. This paper introduces split learning into the two-tower recommendation models and proposes STTFedRec, a privacy-preserving and efficient cross-device federated recommendation framework. STTFedRec achieves local computation reduction by splitting the training and computation of the item model from user devices to a performance-powered server. The server with the item model provides low-dimensional item embeddings instead of raw item data to the user devices for local training and online inferring, achieving server broadcast compression. The user devices only need to perform similarity calculations with cached user embeddings to achieve efficient online inferring. We also propose an obfuscated item request strategy and multi-party circular secret sharing chain to enhance the privacy protection of model training. The experiments conducted on two public datasets demonstrate that STTFedRec improves the average computation time and communication size of the baseline models by about 40 times and 42 times in the best-case scenario with balanced recommendation accuracy." @default.
- W4366196580 created "2023-04-19" @default.
- W4366196580 creator A5004828717 @default.
- W4366196580 creator A5019580763 @default.
- W4366196580 creator A5021881680 @default.
- W4366196580 creator A5030910124 @default.
- W4366196580 date "2023-04-14" @default.
- W4366196580 modified "2023-10-01" @default.
- W4366196580 title "A split-federated learning and edge-cloud based efficient and privacy-preserving large-scale item recommendation model" @default.
- W4366196580 cites W2090615502 @default.
- W4366196580 cites W2125170840 @default.
- W4366196580 cites W2131876387 @default.
- W4366196580 cites W2344423009 @default.
- W4366196580 cites W2626769593 @default.
- W4366196580 cites W2743333519 @default.
- W4366196580 cites W2891123429 @default.
- W4366196580 cites W2895939936 @default.
- W4366196580 cites W2909432978 @default.
- W4366196580 cites W2962814013 @default.
- W4366196580 cites W2963209930 @default.
- W4366196580 cites W2972801466 @default.
- W4366196580 cites W3015636663 @default.
- W4366196580 cites W3023045848 @default.
- W4366196580 cites W3034414418 @default.
- W4366196580 cites W3039440100 @default.
- W4366196580 cites W3047789363 @default.
- W4366196580 cites W3089924157 @default.
- W4366196580 cites W3091870957 @default.
- W4366196580 cites W3102496129 @default.
- W4366196580 cites W3103802018 @default.
- W4366196580 cites W3104789011 @default.
- W4366196580 cites W3134210339 @default.
- W4366196580 cites W3135045645 @default.
- W4366196580 cites W3158146252 @default.
- W4366196580 cites W3181304926 @default.
- W4366196580 cites W3199197205 @default.
- W4366196580 cites W4206326997 @default.
- W4366196580 cites W4281804670 @default.
- W4366196580 cites W4289816670 @default.
- W4366196580 cites W4294106961 @default.
- W4366196580 cites W4312905817 @default.
- W4366196580 doi "https://doi.org/10.1186/s13677-023-00435-5" @default.
- W4366196580 hasPublicationYear "2023" @default.
- W4366196580 type Work @default.
- W4366196580 citedByCount "0" @default.
- W4366196580 crossrefType "journal-article" @default.
- W4366196580 hasAuthorship W4366196580A5004828717 @default.
- W4366196580 hasAuthorship W4366196580A5019580763 @default.
- W4366196580 hasAuthorship W4366196580A5021881680 @default.
- W4366196580 hasAuthorship W4366196580A5030910124 @default.
- W4366196580 hasBestOaLocation W43661965801 @default.
- W4366196580 hasConcept C111919701 @default.
- W4366196580 hasConcept C11413529 @default.
- W4366196580 hasConcept C115537543 @default.
- W4366196580 hasConcept C119857082 @default.
- W4366196580 hasConcept C123201435 @default.
- W4366196580 hasConcept C124101348 @default.
- W4366196580 hasConcept C132964779 @default.
- W4366196580 hasConcept C136764020 @default.
- W4366196580 hasConcept C154945302 @default.
- W4366196580 hasConcept C162307627 @default.
- W4366196580 hasConcept C186967261 @default.
- W4366196580 hasConcept C199360897 @default.
- W4366196580 hasConcept C2778456923 @default.
- W4366196580 hasConcept C2779960059 @default.
- W4366196580 hasConcept C31258907 @default.
- W4366196580 hasConcept C38652104 @default.
- W4366196580 hasConcept C41008148 @default.
- W4366196580 hasConcept C45374587 @default.
- W4366196580 hasConcept C557471498 @default.
- W4366196580 hasConcept C79974875 @default.
- W4366196580 hasConcept C93996380 @default.
- W4366196580 hasConceptScore W4366196580C111919701 @default.
- W4366196580 hasConceptScore W4366196580C11413529 @default.
- W4366196580 hasConceptScore W4366196580C115537543 @default.
- W4366196580 hasConceptScore W4366196580C119857082 @default.
- W4366196580 hasConceptScore W4366196580C123201435 @default.
- W4366196580 hasConceptScore W4366196580C124101348 @default.
- W4366196580 hasConceptScore W4366196580C132964779 @default.
- W4366196580 hasConceptScore W4366196580C136764020 @default.
- W4366196580 hasConceptScore W4366196580C154945302 @default.
- W4366196580 hasConceptScore W4366196580C162307627 @default.
- W4366196580 hasConceptScore W4366196580C186967261 @default.
- W4366196580 hasConceptScore W4366196580C199360897 @default.
- W4366196580 hasConceptScore W4366196580C2778456923 @default.
- W4366196580 hasConceptScore W4366196580C2779960059 @default.
- W4366196580 hasConceptScore W4366196580C31258907 @default.
- W4366196580 hasConceptScore W4366196580C38652104 @default.
- W4366196580 hasConceptScore W4366196580C41008148 @default.
- W4366196580 hasConceptScore W4366196580C45374587 @default.
- W4366196580 hasConceptScore W4366196580C557471498 @default.
- W4366196580 hasConceptScore W4366196580C79974875 @default.
- W4366196580 hasConceptScore W4366196580C93996380 @default.
- W4366196580 hasIssue "1" @default.
- W4366196580 hasLocation W43661965801 @default.
- W4366196580 hasLocation W43661965802 @default.
- W4366196580 hasLocation W43661965803 @default.
- W4366196580 hasOpenAccess W4366196580 @default.
- W4366196580 hasPrimaryLocation W43661965801 @default.
- W4366196580 hasRelatedWork W1583717361 @default.