Matches in SemOpenAlex for { <https://semopenalex.org/work/W3023113746> ?p ?o ?g. }
- W3023113746 abstract "The aim of distributional semantics is to learn the meanings of words from a corpus of text. The aim of formal semantics is to develop mathematical models of meaning. Functional Distributional Semantics provides a framework for distributional semantics which is interpretable in formal semantic terms, by representing the meaning of a word as a truth-conditional function (a binary classifier). However, the model introduces a large number of latent variables, which means that inference is computationally expensive, and training a model is therefore slow to converge. In this work, I introduce the Pixie Autoencoder, which augments the generative model of Functional Distributional Semantics with a graph-convolutional neural network to perform amortised variational inference. This allows the model to be trained more effectively, achieving better results on semantic similarity in context, and outperforming BERT, a large pre-trained language model." @default.
- W3023113746 created "2020-05-13" @default.
- W3023113746 creator A5041079942 @default.
- W3023113746 date "2021-02-02" @default.
- W3023113746 modified "2023-10-16" @default.
- W3023113746 title "Autoencoding Pixies: Amortised Variational Inference with Graph Convolutions for Functional Distributional Semantics" @default.
- W3023113746 cites W138379559 @default.
- W3023113746 cites W1484288670 @default.
- W3023113746 cites W1500718495 @default.
- W3023113746 cites W1501189498 @default.
- W3023113746 cites W1503398984 @default.
- W3023113746 cites W1516111018 @default.
- W3023113746 cites W1548622009 @default.
- W3023113746 cites W1566289585 @default.
- W3023113746 cites W1597195725 @default.
- W3023113746 cites W1614298861 @default.
- W3023113746 cites W168564468 @default.
- W3023113746 cites W1687416839 @default.
- W3023113746 cites W170175252 @default.
- W3023113746 cites W1796264214 @default.
- W3023113746 cites W1880262756 @default.
- W3023113746 cites W1884290075 @default.
- W3023113746 cites W1958547877 @default.
- W3023113746 cites W2003764924 @default.
- W3023113746 cites W2007871210 @default.
- W3023113746 cites W2019599312 @default.
- W3023113746 cites W2025768430 @default.
- W3023113746 cites W2046609227 @default.
- W3023113746 cites W2064148932 @default.
- W3023113746 cites W2084077871 @default.
- W3023113746 cites W2095589793 @default.
- W3023113746 cites W2096871981 @default.
- W3023113746 cites W2097428050 @default.
- W3023113746 cites W2110453538 @default.
- W3023113746 cites W2113517401 @default.
- W3023113746 cites W2136027818 @default.
- W3023113746 cites W2152444902 @default.
- W3023113746 cites W2160971205 @default.
- W3023113746 cites W2165496929 @default.
- W3023113746 cites W2171879392 @default.
- W3023113746 cites W2200913422 @default.
- W3023113746 cites W224572144 @default.
- W3023113746 cites W2250389441 @default.
- W3023113746 cites W2250473257 @default.
- W3023113746 cites W2251294516 @default.
- W3023113746 cites W2277323063 @default.
- W3023113746 cites W2290847742 @default.
- W3023113746 cites W2345905986 @default.
- W3023113746 cites W2408306441 @default.
- W3023113746 cites W2465321302 @default.
- W3023113746 cites W2515879401 @default.
- W3023113746 cites W2525058744 @default.
- W3023113746 cites W2572063200 @default.
- W3023113746 cites W2577255746 @default.
- W3023113746 cites W2589396532 @default.
- W3023113746 cites W2600702321 @default.
- W3023113746 cites W2604497954 @default.
- W3023113746 cites W2606780347 @default.
- W3023113746 cites W2620924815 @default.
- W3023113746 cites W2622551618 @default.
- W3023113746 cites W2752695879 @default.
- W3023113746 cites W2753738274 @default.
- W3023113746 cites W2798935874 @default.
- W3023113746 cites W2798964814 @default.
- W3023113746 cites W2882319491 @default.
- W3023113746 cites W2909561165 @default.
- W3023113746 cites W2953908231 @default.
- W3023113746 cites W2962739339 @default.
- W3023113746 cites W2962813108 @default.
- W3023113746 cites W2962897886 @default.
- W3023113746 cites W2963039614 @default.
- W3023113746 cites W2963223306 @default.
- W3023113746 cites W2963341956 @default.
- W3023113746 cites W2963403868 @default.
- W3023113746 cites W2963679213 @default.
- W3023113746 cites W2964113829 @default.
- W3023113746 cites W2964121744 @default.
- W3023113746 cites W2964195418 @default.
- W3023113746 cites W2964322752 @default.
- W3023113746 cites W2964327327 @default.
- W3023113746 cites W2970476646 @default.
- W3023113746 cites W2979401726 @default.
- W3023113746 cites W3033938264 @default.
- W3023113746 cites W3034806906 @default.
- W3023113746 cites W3214201242 @default.
- W3023113746 cites W593257684 @default.
- W3023113746 cites W832818353 @default.
- W3023113746 doi "https://doi.org/10.33774/coe-2021-1v9x4" @default.
- W3023113746 hasPublicationYear "2021" @default.
- W3023113746 type Work @default.
- W3023113746 sameAs 3023113746 @default.
- W3023113746 citedByCount "0" @default.
- W3023113746 crossrefType "posted-content" @default.
- W3023113746 hasAuthorship W3023113746A5041079942 @default.
- W3023113746 hasBestOaLocation W30231137461 @default.
- W3023113746 hasConcept C101738243 @default.
- W3023113746 hasConcept C130318100 @default.
- W3023113746 hasConcept C132525143 @default.
- W3023113746 hasConcept C154945302 @default.
- W3023113746 hasConcept C184337299 @default.