FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951060897> ?p ?o ?g. }

• W2951060897 abstract "Despite the tremendous empirical success of neural models in natural language processing, many of them lack the strong intuitions that accompany classical machine learning approaches. Recently, connections have been shown between convolutional neural networks (CNNs) and weighted finite state automata (WFSAs), leading to new interpretations and insights. In this work, we show that some recurrent neural networks also share this connection to WFSAs. We characterize this connection formally, defining rational recurrences to be recurrent hidden state update functions that can be written as the Forward calculation of a finite set of WFSAs. We show that several recent neural models use rational recurrences. Our analysis provides a fresh view of these models and facilitates devising new neural architectures that draw inspiration from WFSAs. We present one such model, which performs better than two recent baselines on language modeling and text classification. Our results demonstrate that transferring intuitions from classical models like WFSAs can be an effective approach to designing and understanding neural models." @default.
• W2951060897 created "2019-06-27" @default.
• W2951060897 creator A5007903277 @default.
• W2951060897 creator A5048087523 @default.
• W2951060897 creator A5079854850 @default.
• W2951060897 creator A5088517824 @default.
• W2951060897 date "2018-08-28" @default.
• W2951060897 modified "2023-09-27" @default.
• W2951060897 title "Rational Recurrences" @default.
• W2951060897 cites W1485981043 @default.
• W2951060897 cites W1506437980 @default.
• W2951060897 cites W1591801644 @default.
• W2951060897 cites W1600043458 @default.
• W2951060897 cites W1632114991 @default.
• W2951060897 cites W179875071 @default.
• W2951060897 cites W1908676432 @default.
• W2951060897 cites W1982370770 @default.
• W2951060897 cites W1984151763 @default.
• W2951060897 cites W2007321142 @default.
• W2951060897 cites W2046932483 @default.
• W2951060897 cites W2061873838 @default.
• W2951060897 cites W2064675550 @default.
• W2951060897 cites W2067619114 @default.
• W2951060897 cites W2083889742 @default.
• W2951060897 cites W2105738468 @default.
• W2951060897 cites W2110485445 @default.
• W2951060897 cites W2112251034 @default.
• W2951060897 cites W2112796928 @default.
• W2951060897 cites W2114524997 @default.
• W2951060897 cites W2121553911 @default.
• W2951060897 cites W2123703928 @default.
• W2951060897 cites W2139424692 @default.
• W2951060897 cites W2151159559 @default.
• W2951060897 cites W2157331557 @default.
• W2951060897 cites W2160660844 @default.
• W2951060897 cites W2161195767 @default.
• W2951060897 cites W2167378781 @default.
• W2951060897 cites W2250539671 @default.
• W2951060897 cites W2251939518 @default.
• W2951060897 cites W2254175738 @default.
• W2951060897 cites W2396384435 @default.
• W2951060897 cites W2470595162 @default.
• W2951060897 cites W2553397501 @default.
• W2951060897 cites W2557270725 @default.
• W2951060897 cites W2559554776 @default.
• W2951060897 cites W2563845258 @default.
• W2951060897 cites W2616969219 @default.
• W2951060897 cites W2618101654 @default.
• W2951060897 cites W2741986794 @default.
• W2951060897 cites W2751185861 @default.
• W2951060897 cites W2962731964 @default.
• W2951060897 cites W2962853227 @default.
• W2951060897 cites W2963001778 @default.
• W2951060897 cites W2963042536 @default.
• W2951060897 cites W2963059228 @default.
• W2951060897 cites W2963073938 @default.
• W2951060897 cites W2963174729 @default.
• W2951060897 cites W2963323889 @default.
• W2951060897 cites W2963347649 @default.
• W2951060897 cites W2963374479 @default.
• W2951060897 cites W2963403868 @default.
• W2951060897 cites W2963706817 @default.
• W2951060897 cites W2963748792 @default.
• W2951060897 cites W2963841132 @default.
• W2951060897 cites W2964121744 @default.
• W2951060897 cites W2964263959 @default.
• W2951060897 cites W2964265128 @default.
• W2951060897 cites W2964308564 @default.
• W2951060897 cites W2964347220 @default.
• W2951060897 cites W3125537303 @default.
• W2951060897 cites W2511755179 @default.
• W2951060897 cites W2587931099 @default.
• W2951060897 hasPublicationYear "2018" @default.
• W2951060897 type Work @default.
• W2951060897 sameAs 2951060897 @default.
• W2951060897 citedByCount "1" @default.
• W2951060897 countsByYear W29510608972017 @default.
• W2951060897 crossrefType "posted-content" @default.
• W2951060897 hasAuthorship W2951060897A5007903277 @default.
• W2951060897 hasAuthorship W2951060897A5048087523 @default.
• W2951060897 hasAuthorship W2951060897A5079854850 @default.
• W2951060897 hasAuthorship W2951060897A5088517824 @default.
• W2951060897 hasConcept C112505250 @default.
• W2951060897 hasConcept C11413529 @default.
• W2951060897 hasConcept C119857082 @default.
• W2951060897 hasConcept C13355873 @default.
• W2951060897 hasConcept C147168706 @default.
• W2951060897 hasConcept C154945302 @default.
• W2951060897 hasConcept C167822520 @default.
• W2951060897 hasConcept C177264268 @default.
• W2951060897 hasConcept C199360897 @default.
• W2951060897 hasConcept C2524010 @default.
• W2951060897 hasConcept C33923547 @default.
• W2951060897 hasConcept C41008148 @default.
• W2951060897 hasConcept C50644808 @default.
• W2951060897 hasConcept C80444323 @default.
• W2951060897 hasConcept C81363708 @default.
• W2951060897 hasConceptScore W2951060897C112505250 @default.
• W2951060897 hasConceptScore W2951060897C11413529 @default.
• W2951060897 hasConceptScore W2951060897C119857082 @default.

• next