Matches in SemOpenAlex for { <https://semopenalex.org/work/W4303684103> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4303684103 abstract "Modern recommender systems operate in a fully server-based fashion. To cater to millions of users, the frequent model maintaining and the high-speed processing for concurrent user requests are required, which comes at the cost of a huge carbon footprint. Meanwhile, users need to upload their behavior data even including the immediate environmental context to the server, raising the public concern about privacy. On-device recommender systems circumvent these two issues with cost-conscious settings and local inference. However, due to the limited memory and computing resources, on-device recommender systems are confronted with two fundamental challenges: (1) how to reduce the size of regular models to fit edge devices? (2) how to retain the original capacity? Previous research mostly adopts tensor decomposition techniques to compress the regular recommendation model with limited compression ratio so as to avoid drastic performance degradation. In this paper, we explore ultra-compact models for next-item recommendation, by loosing the constraint of dimensionality consistency in tensor decomposition. Meanwhile, to compensate for the capacity loss caused by compression, we develop a self-supervised knowledge distillation framework which enables the compressed model (student) to distill the essential information lying in the raw data, and improves the long-tail item recommendation through an embedding-recombination strategy with the original model (teacher). The extensive experiments on two benchmarks demonstrate that, with 30x model size reduction, the compressed model almost comes with no accuracy loss, and even outperforms its uncompressed counterpart in most cases." @default.
- W4303684103 created "2022-10-09" @default.
- W4303684103 creator A5005504607 @default.
- W4303684103 creator A5051512158 @default.
- W4303684103 creator A5065191560 @default.
- W4303684103 creator A5069358349 @default.
- W4303684103 creator A5084564297 @default.
- W4303684103 creator A5088492734 @default.
- W4303684103 date "2022-04-23" @default.
- W4303684103 modified "2023-10-17" @default.
- W4303684103 title "On-Device Next-Item Recommendation with Self-Supervised Knowledge Distillation" @default.
- W4303684103 doi "https://doi.org/10.48550/arxiv.2204.11091" @default.
- W4303684103 hasPublicationYear "2022" @default.
- W4303684103 type Work @default.
- W4303684103 citedByCount "0" @default.
- W4303684103 crossrefType "posted-content" @default.
- W4303684103 hasAuthorship W4303684103A5005504607 @default.
- W4303684103 hasAuthorship W4303684103A5051512158 @default.
- W4303684103 hasAuthorship W4303684103A5065191560 @default.
- W4303684103 hasAuthorship W4303684103A5069358349 @default.
- W4303684103 hasAuthorship W4303684103A5084564297 @default.
- W4303684103 hasAuthorship W4303684103A5088492734 @default.
- W4303684103 hasBestOaLocation W43036841031 @default.
- W4303684103 hasConcept C119857082 @default.
- W4303684103 hasConcept C124101348 @default.
- W4303684103 hasConcept C136764020 @default.
- W4303684103 hasConcept C151730666 @default.
- W4303684103 hasConcept C154945302 @default.
- W4303684103 hasConcept C162478608 @default.
- W4303684103 hasConcept C173801870 @default.
- W4303684103 hasConcept C202474056 @default.
- W4303684103 hasConcept C2776214188 @default.
- W4303684103 hasConcept C2779343474 @default.
- W4303684103 hasConcept C2781238097 @default.
- W4303684103 hasConcept C41008148 @default.
- W4303684103 hasConcept C557471498 @default.
- W4303684103 hasConcept C71901391 @default.
- W4303684103 hasConcept C86803240 @default.
- W4303684103 hasConceptScore W4303684103C119857082 @default.
- W4303684103 hasConceptScore W4303684103C124101348 @default.
- W4303684103 hasConceptScore W4303684103C136764020 @default.
- W4303684103 hasConceptScore W4303684103C151730666 @default.
- W4303684103 hasConceptScore W4303684103C154945302 @default.
- W4303684103 hasConceptScore W4303684103C162478608 @default.
- W4303684103 hasConceptScore W4303684103C173801870 @default.
- W4303684103 hasConceptScore W4303684103C202474056 @default.
- W4303684103 hasConceptScore W4303684103C2776214188 @default.
- W4303684103 hasConceptScore W4303684103C2779343474 @default.
- W4303684103 hasConceptScore W4303684103C2781238097 @default.
- W4303684103 hasConceptScore W4303684103C41008148 @default.
- W4303684103 hasConceptScore W4303684103C557471498 @default.
- W4303684103 hasConceptScore W4303684103C71901391 @default.
- W4303684103 hasConceptScore W4303684103C86803240 @default.
- W4303684103 hasLocation W43036841031 @default.
- W4303684103 hasOpenAccess W4303684103 @default.
- W4303684103 hasPrimaryLocation W43036841031 @default.
- W4303684103 hasRelatedWork W1544722666 @default.
- W4303684103 hasRelatedWork W2028415426 @default.
- W4303684103 hasRelatedWork W2348159088 @default.
- W4303684103 hasRelatedWork W2369936857 @default.
- W4303684103 hasRelatedWork W2372117511 @default.
- W4303684103 hasRelatedWork W2809363009 @default.
- W4303684103 hasRelatedWork W2963058055 @default.
- W4303684103 hasRelatedWork W2968745142 @default.
- W4303684103 hasRelatedWork W4280524276 @default.
- W4303684103 hasRelatedWork W4303684103 @default.
- W4303684103 isParatext "false" @default.
- W4303684103 isRetracted "false" @default.
- W4303684103 workType "article" @default.