Matches in SemOpenAlex for { <https://semopenalex.org/work/W3210980860> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W3210980860 abstract "Nowadays, deep learning models are widely adopted in web-scale applications such as recommender systems, and online advertising. In these applications, embedding learning of categorical features is crucial to the success of deep learning models. In these models, a standard method is that each categorical feature value is assigned a unique embedding vector which can be learned and optimized. Although this method can well capture the characteristics of the categorical features and promise good performance, it can incur a huge memory cost to store the embedding table, especially for those web-scale applications. Such a huge memory cost significantly holds back the effectiveness and usability of EDRMs. In this paper, we propose a binary code based hash embedding method which allows the size of the embedding table to be reduced in arbitrary scale without compromising too much performance. Experimental evaluation results show that one can still achieve 99% performance even if the embedding table size is reduced 1000× smaller than the original one with our proposed method." @default.
- W3210980860 created "2021-11-08" @default.
- W3210980860 creator A5004319733 @default.
- W3210980860 creator A5007939424 @default.
- W3210980860 creator A5017996548 @default.
- W3210980860 creator A5028779371 @default.
- W3210980860 creator A5040706602 @default.
- W3210980860 creator A5062270059 @default.
- W3210980860 creator A5073856221 @default.
- W3210980860 date "2021-10-26" @default.
- W3210980860 modified "2023-09-25" @default.
- W3210980860 title "Binary Code based Hash Embedding for Web-scale Applications" @default.
- W3210980860 cites W1967021753 @default.
- W3210980860 cites W2070996757 @default.
- W3210980860 cites W2077815765 @default.
- W3210980860 cites W2124509324 @default.
- W3210980860 cites W2158698691 @default.
- W3210980860 cites W2475334473 @default.
- W3210980860 cites W2626376796 @default.
- W3210980860 cites W2723293840 @default.
- W3210980860 cites W2755721434 @default.
- W3210980860 cites W2964182926 @default.
- W3210980860 cites W2983088090 @default.
- W3210980860 cites W3023259715 @default.
- W3210980860 cites W3034552531 @default.
- W3210980860 cites W3081362488 @default.
- W3210980860 cites W3088463703 @default.
- W3210980860 cites W3101704389 @default.
- W3210980860 doi "https://doi.org/10.1145/3459637.3482065" @default.
- W3210980860 hasPublicationYear "2021" @default.
- W3210980860 type Work @default.
- W3210980860 sameAs 3210980860 @default.
- W3210980860 citedByCount "2" @default.
- W3210980860 countsByYear W32109808602023 @default.
- W3210980860 crossrefType "proceedings-article" @default.
- W3210980860 hasAuthorship W3210980860A5004319733 @default.
- W3210980860 hasAuthorship W3210980860A5007939424 @default.
- W3210980860 hasAuthorship W3210980860A5017996548 @default.
- W3210980860 hasAuthorship W3210980860A5028779371 @default.
- W3210980860 hasAuthorship W3210980860A5040706602 @default.
- W3210980860 hasAuthorship W3210980860A5062270059 @default.
- W3210980860 hasAuthorship W3210980860A5073856221 @default.
- W3210980860 hasBestOaLocation W32109808602 @default.
- W3210980860 hasConcept C107457646 @default.
- W3210980860 hasConcept C108583219 @default.
- W3210980860 hasConcept C119857082 @default.
- W3210980860 hasConcept C121332964 @default.
- W3210980860 hasConcept C124101348 @default.
- W3210980860 hasConcept C154945302 @default.
- W3210980860 hasConcept C170130773 @default.
- W3210980860 hasConcept C177264268 @default.
- W3210980860 hasConcept C199360897 @default.
- W3210980860 hasConcept C2776760102 @default.
- W3210980860 hasConcept C2778755073 @default.
- W3210980860 hasConcept C41008148 @default.
- W3210980860 hasConcept C41608201 @default.
- W3210980860 hasConcept C45235069 @default.
- W3210980860 hasConcept C5274069 @default.
- W3210980860 hasConcept C62520636 @default.
- W3210980860 hasConcept C67388219 @default.
- W3210980860 hasConcept C80444323 @default.
- W3210980860 hasConcept C99138194 @default.
- W3210980860 hasConceptScore W3210980860C107457646 @default.
- W3210980860 hasConceptScore W3210980860C108583219 @default.
- W3210980860 hasConceptScore W3210980860C119857082 @default.
- W3210980860 hasConceptScore W3210980860C121332964 @default.
- W3210980860 hasConceptScore W3210980860C124101348 @default.
- W3210980860 hasConceptScore W3210980860C154945302 @default.
- W3210980860 hasConceptScore W3210980860C170130773 @default.
- W3210980860 hasConceptScore W3210980860C177264268 @default.
- W3210980860 hasConceptScore W3210980860C199360897 @default.
- W3210980860 hasConceptScore W3210980860C2776760102 @default.
- W3210980860 hasConceptScore W3210980860C2778755073 @default.
- W3210980860 hasConceptScore W3210980860C41008148 @default.
- W3210980860 hasConceptScore W3210980860C41608201 @default.
- W3210980860 hasConceptScore W3210980860C45235069 @default.
- W3210980860 hasConceptScore W3210980860C5274069 @default.
- W3210980860 hasConceptScore W3210980860C62520636 @default.
- W3210980860 hasConceptScore W3210980860C67388219 @default.
- W3210980860 hasConceptScore W3210980860C80444323 @default.
- W3210980860 hasConceptScore W3210980860C99138194 @default.
- W3210980860 hasLocation W32109808601 @default.
- W3210980860 hasLocation W32109808602 @default.
- W3210980860 hasOpenAccess W3210980860 @default.
- W3210980860 hasPrimaryLocation W32109808601 @default.
- W3210980860 hasRelatedWork W3014300295 @default.
- W3210980860 hasRelatedWork W3164822677 @default.
- W3210980860 hasRelatedWork W4223943233 @default.
- W3210980860 hasRelatedWork W4225161397 @default.
- W3210980860 hasRelatedWork W4250304930 @default.
- W3210980860 hasRelatedWork W4312200629 @default.
- W3210980860 hasRelatedWork W4360585206 @default.
- W3210980860 hasRelatedWork W4364306694 @default.
- W3210980860 hasRelatedWork W4380075502 @default.
- W3210980860 hasRelatedWork W4380086463 @default.
- W3210980860 isParatext "false" @default.
- W3210980860 isRetracted "false" @default.
- W3210980860 magId "3210980860" @default.
- W3210980860 workType "article" @default.