FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2894844282> ?p ?o ?g. }

• W2894844282 abstract "In order to build efficient deep recurrent neural architectures, it is essential to analyze the complexityof long distance dependencies (LDDs) of the dataset being modeled. In this paper, we presentdetailed analysis of the dependency decay curve exhibited by various datasets. The datasets sampledfrom a similar process (e.g. natural language, sequential MNIST, Strictlyk-Piecewise languages,etc) display variations in the properties of the dependency decay curve. Our analysis reveal thefactors resulting in these variations; such as (i) number of unique symbols in a dataset, (ii) size ofthe dataset, (iii) number of interacting symbols within a given LDD, and (iv) the distance betweenthe interacting symbols. We test these factors by generating synthesized datasets of the Strictlyk-Piecewise languages. Another advantage of these synthesized datasets is that they enable targetedtesting of deep recurrent neural architectures in terms of their ability to model LDDs with differentcharacteristics. We also demonstrate that analysing dependency decay curves can inform the selectionof optimal hyper-parameters for SOTA deep recurrent neural architectures. This analysis can directlycontribute to the development of more accurate and efficient sequential models." @default.
• W2894844282 created "2018-10-12" @default.
• W2894844282 creator A5058781219 @default.
• W2894844282 creator A5086319818 @default.
• W2894844282 date "2018-10-06" @default.
• W2894844282 modified "2023-09-27" @default.
• W2894844282 title "Understanding Recurrent Neural Architectures by Analyzing and Synthesizing Long Distance Dependencies in Benchmark Sequential Datasets." @default.
• W2894844282 cites W1498437823 @default.
• W2894844282 cites W1533338086 @default.
• W2894844282 cites W1543949261 @default.
• W2894844282 cites W1661915592 @default.
• W2894844282 cites W1662195770 @default.
• W2894844282 cites W1966547129 @default.
• W2894844282 cites W1982370770 @default.
• W2894844282 cites W1992020520 @default.
• W2894844282 cites W2012796055 @default.
• W2894844282 cites W2040713190 @default.
• W2894844282 cites W2057953526 @default.
• W2894844282 cites W2064675550 @default.
• W2894844282 cites W2066873261 @default.
• W2894844282 cites W2081965739 @default.
• W2894844282 cites W2082345008 @default.
• W2894844282 cites W2093647425 @default.
• W2894844282 cites W2095881968 @default.
• W2894844282 cites W2099111195 @default.
• W2894844282 cites W2099257174 @default.
• W2894844282 cites W2107878631 @default.
• W2894844282 cites W2108789285 @default.
• W2894844282 cites W2110485445 @default.
• W2894844282 cites W2114771311 @default.
• W2894844282 cites W2115912164 @default.
• W2894844282 cites W2124479173 @default.
• W2894844282 cites W2146305921 @default.
• W2894844282 cites W2525332836 @default.
• W2894844282 cites W2616724617 @default.
• W2894844282 cites W2775624460 @default.
• W2894844282 cites W2803437449 @default.
• W2894844282 cites W2888799392 @default.
• W2894844282 cites W2950527759 @default.
• W2894844282 cites W2962832505 @default.
• W2894844282 cites W2962883166 @default.
• W2894844282 cites W2963241221 @default.
• W2894844282 cites W2963403868 @default.
• W2894844282 cites W2963430354 @default.
• W2894844282 cites W2963537482 @default.
• W2894844282 cites W2963631907 @default.
• W2894844282 cites W2963938518 @default.
• W2894844282 cites W2964182247 @default.
• W2894844282 cites W2993383518 @default.
• W2894844282 cites W3099850795 @default.
• W2894844282 cites W3102543255 @default.
• W2894844282 cites W3105376354 @default.
• W2894844282 cites W1525783482 @default.
• W2894844282 cites W194249466 @default.
• W2894844282 hasPublicationYear "2018" @default.
• W2894844282 type Work @default.
• W2894844282 sameAs 2894844282 @default.
• W2894844282 citedByCount "6" @default.
• W2894844282 countsByYear W28948442822018 @default.
• W2894844282 countsByYear W28948442822019 @default.
• W2894844282 countsByYear W28948442822020 @default.
• W2894844282 crossrefType "posted-content" @default.
• W2894844282 hasAuthorship W2894844282A5058781219 @default.
• W2894844282 hasAuthorship W2894844282A5086319818 @default.
• W2894844282 hasConcept C108583219 @default.
• W2894844282 hasConcept C111919701 @default.
• W2894844282 hasConcept C11413529 @default.
• W2894844282 hasConcept C124101348 @default.
• W2894844282 hasConcept C13280743 @default.
• W2894844282 hasConcept C134306372 @default.
• W2894844282 hasConcept C147168706 @default.
• W2894844282 hasConcept C153180895 @default.
• W2894844282 hasConcept C154945302 @default.
• W2894844282 hasConcept C164660894 @default.
• W2894844282 hasConcept C185798385 @default.
• W2894844282 hasConcept C190502265 @default.
• W2894844282 hasConcept C19768560 @default.
• W2894844282 hasConcept C205649164 @default.
• W2894844282 hasConcept C33923547 @default.
• W2894844282 hasConcept C41008148 @default.
• W2894844282 hasConcept C50644808 @default.
• W2894844282 hasConcept C98045186 @default.
• W2894844282 hasConceptScore W2894844282C108583219 @default.
• W2894844282 hasConceptScore W2894844282C111919701 @default.
• W2894844282 hasConceptScore W2894844282C11413529 @default.
• W2894844282 hasConceptScore W2894844282C124101348 @default.
• W2894844282 hasConceptScore W2894844282C13280743 @default.
• W2894844282 hasConceptScore W2894844282C134306372 @default.
• W2894844282 hasConceptScore W2894844282C147168706 @default.
• W2894844282 hasConceptScore W2894844282C153180895 @default.
• W2894844282 hasConceptScore W2894844282C154945302 @default.
• W2894844282 hasConceptScore W2894844282C164660894 @default.
• W2894844282 hasConceptScore W2894844282C185798385 @default.
• W2894844282 hasConceptScore W2894844282C190502265 @default.
• W2894844282 hasConceptScore W2894844282C19768560 @default.
• W2894844282 hasConceptScore W2894844282C205649164 @default.
• W2894844282 hasConceptScore W2894844282C33923547 @default.
• W2894844282 hasConceptScore W2894844282C41008148 @default.
• W2894844282 hasConceptScore W2894844282C50644808 @default.
• W2894844282 hasConceptScore W2894844282C98045186 @default.

• next