Matches in SemOpenAlex for { <https://semopenalex.org/work/W4221145950> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W4221145950 abstract "Modeling long range dependencies in sequential data is a fundamental step towards attaining human-level performance in many modalities such as text, vision, audio and video. While attention-based models are a popular and effective choice in modeling short-range interactions, their performance on tasks requiring long range reasoning has been largely inadequate. In an exciting result, Gu et al. (ICLR 2022) proposed the $textit{Structured State Space}$ (S4) architecture delivering large gains over state-of-the-art models on several long-range tasks across various modalities. The core proposition of S4 is the parameterization of state matrices via a diagonal plus low rank structure, allowing efficient computation. In this work, we show that one can match the performance of S4 even without the low rank correction and thus assuming the state matrices to be diagonal. Our $textit{Diagonal State Space}$ (DSS) model matches the performance of S4 on Long Range Arena tasks, speech classification on Speech Commands dataset, while being conceptually simpler and straightforward to implement." @default.
- W4221145950 created "2022-04-03" @default.
- W4221145950 creator A5022734062 @default.
- W4221145950 date "2022-03-27" @default.
- W4221145950 modified "2023-10-14" @default.
- W4221145950 title "Diagonal State Spaces are as Effective as Structured State Spaces" @default.
- W4221145950 doi "https://doi.org/10.48550/arxiv.2203.14343" @default.
- W4221145950 hasPublicationYear "2022" @default.
- W4221145950 type Work @default.
- W4221145950 citedByCount "0" @default.
- W4221145950 crossrefType "posted-content" @default.
- W4221145950 hasAuthorship W4221145950A5022734062 @default.
- W4221145950 hasBestOaLocation W42211459501 @default.
- W4221145950 hasConcept C105795698 @default.
- W4221145950 hasConcept C106487976 @default.
- W4221145950 hasConcept C111919701 @default.
- W4221145950 hasConcept C11413529 @default.
- W4221145950 hasConcept C114614502 @default.
- W4221145950 hasConcept C130367717 @default.
- W4221145950 hasConcept C138885662 @default.
- W4221145950 hasConcept C144024400 @default.
- W4221145950 hasConcept C154945302 @default.
- W4221145950 hasConcept C159985019 @default.
- W4221145950 hasConcept C164226766 @default.
- W4221145950 hasConcept C173608175 @default.
- W4221145950 hasConcept C192562407 @default.
- W4221145950 hasConcept C204323151 @default.
- W4221145950 hasConcept C2524010 @default.
- W4221145950 hasConcept C2777152325 @default.
- W4221145950 hasConcept C2778572836 @default.
- W4221145950 hasConcept C2779903281 @default.
- W4221145950 hasConcept C33923547 @default.
- W4221145950 hasConcept C36289849 @default.
- W4221145950 hasConcept C3826847 @default.
- W4221145950 hasConcept C41008148 @default.
- W4221145950 hasConcept C41895202 @default.
- W4221145950 hasConcept C45374587 @default.
- W4221145950 hasConcept C48103436 @default.
- W4221145950 hasConcept C72434380 @default.
- W4221145950 hasConcept C80444323 @default.
- W4221145950 hasConceptScore W4221145950C105795698 @default.
- W4221145950 hasConceptScore W4221145950C106487976 @default.
- W4221145950 hasConceptScore W4221145950C111919701 @default.
- W4221145950 hasConceptScore W4221145950C11413529 @default.
- W4221145950 hasConceptScore W4221145950C114614502 @default.
- W4221145950 hasConceptScore W4221145950C130367717 @default.
- W4221145950 hasConceptScore W4221145950C138885662 @default.
- W4221145950 hasConceptScore W4221145950C144024400 @default.
- W4221145950 hasConceptScore W4221145950C154945302 @default.
- W4221145950 hasConceptScore W4221145950C159985019 @default.
- W4221145950 hasConceptScore W4221145950C164226766 @default.
- W4221145950 hasConceptScore W4221145950C173608175 @default.
- W4221145950 hasConceptScore W4221145950C192562407 @default.
- W4221145950 hasConceptScore W4221145950C204323151 @default.
- W4221145950 hasConceptScore W4221145950C2524010 @default.
- W4221145950 hasConceptScore W4221145950C2777152325 @default.
- W4221145950 hasConceptScore W4221145950C2778572836 @default.
- W4221145950 hasConceptScore W4221145950C2779903281 @default.
- W4221145950 hasConceptScore W4221145950C33923547 @default.
- W4221145950 hasConceptScore W4221145950C36289849 @default.
- W4221145950 hasConceptScore W4221145950C3826847 @default.
- W4221145950 hasConceptScore W4221145950C41008148 @default.
- W4221145950 hasConceptScore W4221145950C41895202 @default.
- W4221145950 hasConceptScore W4221145950C45374587 @default.
- W4221145950 hasConceptScore W4221145950C48103436 @default.
- W4221145950 hasConceptScore W4221145950C72434380 @default.
- W4221145950 hasConceptScore W4221145950C80444323 @default.
- W4221145950 hasLocation W42211459501 @default.
- W4221145950 hasLocation W42211459502 @default.
- W4221145950 hasOpenAccess W4221145950 @default.
- W4221145950 hasPrimaryLocation W42211459501 @default.
- W4221145950 hasRelatedWork W1524943583 @default.
- W4221145950 hasRelatedWork W1881102326 @default.
- W4221145950 hasRelatedWork W1980294706 @default.
- W4221145950 hasRelatedWork W2085477326 @default.
- W4221145950 hasRelatedWork W2105730801 @default.
- W4221145950 hasRelatedWork W2326195745 @default.
- W4221145950 hasRelatedWork W2384229502 @default.
- W4221145950 hasRelatedWork W3013553788 @default.
- W4221145950 hasRelatedWork W3097836484 @default.
- W4221145950 hasRelatedWork W4323640762 @default.
- W4221145950 isParatext "false" @default.
- W4221145950 isRetracted "false" @default.
- W4221145950 workType "article" @default.