Matches in SemOpenAlex for { <https://semopenalex.org/work/W2530887700> ?p ?o ?g. }
- W2530887700 endingPage "476" @default.
- W2530887700 startingPage "471" @default.
- W2530887700 abstract "Artificial neural networks are remarkably adept at sensory processing, sequence learning and reinforcement learning, but are limited in their ability to represent variables and data structures and to store data over long timescales, owing to the lack of an external memory. Here we introduce a machine learning model called a differentiable neural computer (DNC), which consists of a neural network that can read from and write to an external memory matrix, analogous to the random-access memory in a conventional computer. Like a conventional computer, it can use its memory to represent and manipulate complex data structures, but, like a neural network, it can learn to do so from data. When trained with supervised learning, we demonstrate that a DNC can successfully answer synthetic questions designed to emulate reasoning and inference problems in natural language. We show that it can learn tasks such as finding the shortest path between specified points and inferring the missing links in randomly generated graphs, and then generalize these tasks to specific graphs such as transport networks and family trees. When trained with reinforcement learning, a DNC can complete a moving blocks puzzle in which changing goals are specified by sequences of symbols. Taken together, our results demonstrate that DNCs have the capacity to solve complex, structured tasks that are inaccessible to neural networks without external read-write memory." @default.
- W2530887700 created "2016-10-21" @default.
- W2530887700 creator A5001284503 @default.
- W2530887700 creator A5002017252 @default.
- W2530887700 creator A5005349213 @default.
- W2530887700 creator A5013028446 @default.
- W2530887700 creator A5014309771 @default.
- W2530887700 creator A5016618355 @default.
- W2530887700 creator A5016838762 @default.
- W2530887700 creator A5023301183 @default.
- W2530887700 creator A5023508792 @default.
- W2530887700 creator A5026921940 @default.
- W2530887700 creator A5031878516 @default.
- W2530887700 creator A5038813361 @default.
- W2530887700 creator A5043473089 @default.
- W2530887700 creator A5048371195 @default.
- W2530887700 creator A5059282957 @default.
- W2530887700 creator A5059552797 @default.
- W2530887700 creator A5073651612 @default.
- W2530887700 creator A5074976302 @default.
- W2530887700 creator A5089497713 @default.
- W2530887700 creator A5090341705 @default.
- W2530887700 date "2016-10-01" @default.
- W2530887700 modified "2023-10-16" @default.
- W2530887700 title "Hybrid computing using a neural network with dynamic external memory" @default.
- W2530887700 cites W1899349177 @default.
- W2530887700 cites W1997123519 @default.
- W2530887700 cites W2001788447 @default.
- W2530887700 cites W2024418895 @default.
- W2530887700 cites W2047057213 @default.
- W2530887700 cites W2051459061 @default.
- W2530887700 cites W2064675550 @default.
- W2530887700 cites W2072233684 @default.
- W2530887700 cites W2095739681 @default.
- W2530887700 cites W2105398397 @default.
- W2530887700 cites W2107139863 @default.
- W2530887700 cites W2120629471 @default.
- W2530887700 cites W2129386380 @default.
- W2530887700 cites W2143612262 @default.
- W2530887700 cites W2145339207 @default.
- W2530887700 cites W2150355110 @default.
- W2530887700 cites W2158349948 @default.
- W2530887700 cites W2159564241 @default.
- W2530887700 cites W2194319389 @default.
- W2530887700 cites W2194321275 @default.
- W2530887700 cites W2296073425 @default.
- W2530887700 cites W2424347275 @default.
- W2530887700 cites W4235344441 @default.
- W2530887700 cites W4247601884 @default.
- W2530887700 doi "https://doi.org/10.1038/nature20101" @default.
- W2530887700 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/27732574" @default.
- W2530887700 hasPublicationYear "2016" @default.
- W2530887700 type Work @default.
- W2530887700 sameAs 2530887700 @default.
- W2530887700 citedByCount "1013" @default.
- W2530887700 countsByYear W25308877002016 @default.
- W2530887700 countsByYear W25308877002017 @default.
- W2530887700 countsByYear W25308877002018 @default.
- W2530887700 countsByYear W25308877002019 @default.
- W2530887700 countsByYear W25308877002020 @default.
- W2530887700 countsByYear W25308877002021 @default.
- W2530887700 countsByYear W25308877002022 @default.
- W2530887700 countsByYear W25308877002023 @default.
- W2530887700 crossrefType "journal-article" @default.
- W2530887700 hasAuthorship W2530887700A5001284503 @default.
- W2530887700 hasAuthorship W2530887700A5002017252 @default.
- W2530887700 hasAuthorship W2530887700A5005349213 @default.
- W2530887700 hasAuthorship W2530887700A5013028446 @default.
- W2530887700 hasAuthorship W2530887700A5014309771 @default.
- W2530887700 hasAuthorship W2530887700A5016618355 @default.
- W2530887700 hasAuthorship W2530887700A5016838762 @default.
- W2530887700 hasAuthorship W2530887700A5023301183 @default.
- W2530887700 hasAuthorship W2530887700A5023508792 @default.
- W2530887700 hasAuthorship W2530887700A5026921940 @default.
- W2530887700 hasAuthorship W2530887700A5031878516 @default.
- W2530887700 hasAuthorship W2530887700A5038813361 @default.
- W2530887700 hasAuthorship W2530887700A5043473089 @default.
- W2530887700 hasAuthorship W2530887700A5048371195 @default.
- W2530887700 hasAuthorship W2530887700A5059282957 @default.
- W2530887700 hasAuthorship W2530887700A5059552797 @default.
- W2530887700 hasAuthorship W2530887700A5073651612 @default.
- W2530887700 hasAuthorship W2530887700A5074976302 @default.
- W2530887700 hasAuthorship W2530887700A5089497713 @default.
- W2530887700 hasAuthorship W2530887700A5090341705 @default.
- W2530887700 hasBestOaLocation W25308877002 @default.
- W2530887700 hasConcept C119857082 @default.
- W2530887700 hasConcept C147168706 @default.
- W2530887700 hasConcept C154945302 @default.
- W2530887700 hasConcept C2776214188 @default.
- W2530887700 hasConcept C41008148 @default.
- W2530887700 hasConcept C50644808 @default.
- W2530887700 hasConcept C80444323 @default.
- W2530887700 hasConcept C82687282 @default.
- W2530887700 hasConcept C9390403 @default.
- W2530887700 hasConcept C97541855 @default.
- W2530887700 hasConceptScore W2530887700C119857082 @default.
- W2530887700 hasConceptScore W2530887700C147168706 @default.
- W2530887700 hasConceptScore W2530887700C154945302 @default.