Matches in SemOpenAlex for { <https://semopenalex.org/work/W3214702976> ?p ?o ?g. }
- W3214702976 endingPage "733" @default.
- W3214702976 startingPage "719" @default.
- W3214702976 abstract "In this paper, we explore the task of automatically generating natural language descriptions of salient patterns in a time series, such as stock prices of a company over a week. A model for this task should be able to extract high-level patterns such as presence of a peak or a dip. While typical contemporary neural models with attention mechanisms can generate fluent output descriptions for this task, they often generate factually incorrect descriptions. We propose a computational model with a truth-conditional architecture which first runs small learned programs on the input time series, then identifies the programs/patterns which hold true for the given input, and finally conditions on *only* the chosen valid program (rather than the input time series) to generate the output text description. A program in our model is constructed from modules, which are small neural networks that are designed to capture numerical patterns and temporal information. The modules are shared across multiple programs, enabling compositionality as well as efficient learning of module parameters. The modules, as well as the composition of the modules, are unobserved in data, and we learn them in an end-to-end fashion with the only training signal coming from the accompanying natural language text descriptions. We find that the proposed model is able to generate high-precision captions even though we consider a small and simple space of module types." @default.
- W3214702976 created "2021-11-22" @default.
- W3214702976 creator A5015256182 @default.
- W3214702976 creator A5017455302 @default.
- W3214702976 date "2021-11-01" @default.
- W3214702976 modified "2023-09-25" @default.
- W3214702976 title "Truth-Conditional Captions for Time Series Data" @default.
- W3214702976 cites W2101105183 @default.
- W3214702976 cites W2123301721 @default.
- W3214702976 cites W2154652894 @default.
- W3214702976 cites W2250318122 @default.
- W3214702976 cites W2739046565 @default.
- W3214702976 cites W2741690481 @default.
- W3214702976 cites W2767894523 @default.
- W3214702976 cites W2798273548 @default.
- W3214702976 cites W2799225444 @default.
- W3214702976 cites W2887020936 @default.
- W3214702976 cites W2889518897 @default.
- W3214702976 cites W2919420119 @default.
- W3214702976 cites W2948556812 @default.
- W3214702976 cites W2952523122 @default.
- W3214702976 cites W2962700105 @default.
- W3214702976 cites W2962844293 @default.
- W3214702976 cites W2963091658 @default.
- W3214702976 cites W2963143606 @default.
- W3214702976 cites W2963202404 @default.
- W3214702976 cites W2963411289 @default.
- W3214702976 cites W2963676814 @default.
- W3214702976 cites W2963799213 @default.
- W3214702976 cites W2964118342 @default.
- W3214702976 cites W2964196083 @default.
- W3214702976 cites W2970609357 @default.
- W3214702976 cites W2981259322 @default.
- W3214702976 cites W2990069284 @default.
- W3214702976 cites W2995649781 @default.
- W3214702976 cites W2996403597 @default.
- W3214702976 cites W3022814719 @default.
- W3214702976 cites W3034556525 @default.
- W3214702976 cites W3035393139 @default.
- W3214702976 cites W3035725000 @default.
- W3214702976 cites W3037013468 @default.
- W3214702976 cites W3105643199 @default.
- W3214702976 cites W3166890286 @default.
- W3214702976 cites W3169283369 @default.
- W3214702976 cites W3172120146 @default.
- W3214702976 cites W3172267148 @default.
- W3214702976 cites W3175929660 @default.
- W3214702976 cites W3186655327 @default.
- W3214702976 cites W813301878 @default.
- W3214702976 hasPublicationYear "2021" @default.
- W3214702976 type Work @default.
- W3214702976 sameAs 3214702976 @default.
- W3214702976 citedByCount "0" @default.
- W3214702976 crossrefType "proceedings-article" @default.
- W3214702976 hasAuthorship W3214702976A5015256182 @default.
- W3214702976 hasAuthorship W3214702976A5017455302 @default.
- W3214702976 hasConcept C119857082 @default.
- W3214702976 hasConcept C121375916 @default.
- W3214702976 hasConcept C137293760 @default.
- W3214702976 hasConcept C143724316 @default.
- W3214702976 hasConcept C151406439 @default.
- W3214702976 hasConcept C151730666 @default.
- W3214702976 hasConcept C154945302 @default.
- W3214702976 hasConcept C162324750 @default.
- W3214702976 hasConcept C187736073 @default.
- W3214702976 hasConcept C195324797 @default.
- W3214702976 hasConcept C204321447 @default.
- W3214702976 hasConcept C2780451532 @default.
- W3214702976 hasConcept C2780719617 @default.
- W3214702976 hasConcept C41008148 @default.
- W3214702976 hasConcept C86803240 @default.
- W3214702976 hasConceptScore W3214702976C119857082 @default.
- W3214702976 hasConceptScore W3214702976C121375916 @default.
- W3214702976 hasConceptScore W3214702976C137293760 @default.
- W3214702976 hasConceptScore W3214702976C143724316 @default.
- W3214702976 hasConceptScore W3214702976C151406439 @default.
- W3214702976 hasConceptScore W3214702976C151730666 @default.
- W3214702976 hasConceptScore W3214702976C154945302 @default.
- W3214702976 hasConceptScore W3214702976C162324750 @default.
- W3214702976 hasConceptScore W3214702976C187736073 @default.
- W3214702976 hasConceptScore W3214702976C195324797 @default.
- W3214702976 hasConceptScore W3214702976C204321447 @default.
- W3214702976 hasConceptScore W3214702976C2780451532 @default.
- W3214702976 hasConceptScore W3214702976C2780719617 @default.
- W3214702976 hasConceptScore W3214702976C41008148 @default.
- W3214702976 hasConceptScore W3214702976C86803240 @default.
- W3214702976 hasLocation W32147029761 @default.
- W3214702976 hasOpenAccess W3214702976 @default.
- W3214702976 hasPrimaryLocation W32147029761 @default.
- W3214702976 hasRelatedWork W1572851248 @default.
- W3214702976 hasRelatedWork W1780856595 @default.
- W3214702976 hasRelatedWork W2137251014 @default.
- W3214702976 hasRelatedWork W2250318122 @default.
- W3214702976 hasRelatedWork W2592676209 @default.
- W3214702976 hasRelatedWork W2747779981 @default.
- W3214702976 hasRelatedWork W2769134508 @default.
- W3214702976 hasRelatedWork W2796779706 @default.
- W3214702976 hasRelatedWork W2888693386 @default.