Matches in SemOpenAlex for { <https://semopenalex.org/work/W3040607105> ?p ?o ?g. }
- W3040607105 abstract "We examine the necessary and sufficient complexity of neural networks to approximate functions from different smoothness spaces under the restriction of encodable network weights. Based on an entropy argument, we start by proving lower bounds for the number of nonzero encodable weights for neural network approximation in Besov spaces, Sobolev spaces and more. These results are valid for all sufficiently smooth activation functions. Afterwards, we provide a unifying framework for the construction of approximate partitions of unity by neural networks with fairly general activation functions. This allows us to approximate localized Taylor polynomials by neural networks and make use of the Bramble-Hilbert Lemma. Based on our framework, we derive almost optimal upper bounds in higher-order Sobolev norms. This work advances the theory of approximating solutions of partial differential equations by neural networks." @default.
- W3040607105 created "2020-07-10" @default.
- W3040607105 creator A5008776954 @default.
- W3040607105 creator A5067342276 @default.
- W3040607105 date "2020-06-30" @default.
- W3040607105 modified "2023-10-01" @default.
- W3040607105 title "Approximation Rates for Neural Networks with Encodable Weights in Smoothness Spaces" @default.
- W3040607105 cites W1986164709 @default.
- W3040607105 cites W1988115241 @default.
- W3040607105 cites W2019022049 @default.
- W3040607105 cites W2079224763 @default.
- W3040607105 cites W2103496339 @default.
- W3040607105 cites W2152593035 @default.
- W3040607105 cites W2158581396 @default.
- W3040607105 cites W2176412452 @default.
- W3040607105 cites W2528305538 @default.
- W3040607105 cites W2616463902 @default.
- W3040607105 cites W2760972773 @default.
- W3040607105 cites W2787302432 @default.
- W3040607105 cites W2805915852 @default.
- W3040607105 cites W2883486956 @default.
- W3040607105 cites W2890291741 @default.
- W3040607105 cites W2890889625 @default.
- W3040607105 cites W2898740753 @default.
- W3040607105 cites W2933916181 @default.
- W3040607105 cites W2934526567 @default.
- W3040607105 cites W2945744995 @default.
- W3040607105 cites W2952226704 @default.
- W3040607105 cites W2962761333 @default.
- W3040607105 cites W2963146412 @default.
- W3040607105 cites W2963482148 @default.
- W3040607105 cites W2987232971 @default.
- W3040607105 cites W2998882635 @default.
- W3040607105 cites W3020191331 @default.
- W3040607105 cites W3025750359 @default.
- W3040607105 cites W3099056690 @default.
- W3040607105 cites W3101333873 @default.
- W3040607105 hasPublicationYear "2020" @default.
- W3040607105 type Work @default.
- W3040607105 sameAs 3040607105 @default.
- W3040607105 citedByCount "1" @default.
- W3040607105 countsByYear W30406071052021 @default.
- W3040607105 crossrefType "posted-content" @default.
- W3040607105 hasAuthorship W3040607105A5008776954 @default.
- W3040607105 hasAuthorship W3040607105A5067342276 @default.
- W3040607105 hasConcept C102634674 @default.
- W3040607105 hasConcept C106301342 @default.
- W3040607105 hasConcept C118615104 @default.
- W3040607105 hasConcept C121332964 @default.
- W3040607105 hasConcept C134306372 @default.
- W3040607105 hasConcept C154945302 @default.
- W3040607105 hasConcept C18903297 @default.
- W3040607105 hasConcept C202444582 @default.
- W3040607105 hasConcept C2777759810 @default.
- W3040607105 hasConcept C28826006 @default.
- W3040607105 hasConcept C33923547 @default.
- W3040607105 hasConcept C41008148 @default.
- W3040607105 hasConcept C46757340 @default.
- W3040607105 hasConcept C50644808 @default.
- W3040607105 hasConcept C62520636 @default.
- W3040607105 hasConcept C62799726 @default.
- W3040607105 hasConcept C86803240 @default.
- W3040607105 hasConcept C99730327 @default.
- W3040607105 hasConceptScore W3040607105C102634674 @default.
- W3040607105 hasConceptScore W3040607105C106301342 @default.
- W3040607105 hasConceptScore W3040607105C118615104 @default.
- W3040607105 hasConceptScore W3040607105C121332964 @default.
- W3040607105 hasConceptScore W3040607105C134306372 @default.
- W3040607105 hasConceptScore W3040607105C154945302 @default.
- W3040607105 hasConceptScore W3040607105C18903297 @default.
- W3040607105 hasConceptScore W3040607105C202444582 @default.
- W3040607105 hasConceptScore W3040607105C2777759810 @default.
- W3040607105 hasConceptScore W3040607105C28826006 @default.
- W3040607105 hasConceptScore W3040607105C33923547 @default.
- W3040607105 hasConceptScore W3040607105C41008148 @default.
- W3040607105 hasConceptScore W3040607105C46757340 @default.
- W3040607105 hasConceptScore W3040607105C50644808 @default.
- W3040607105 hasConceptScore W3040607105C62520636 @default.
- W3040607105 hasConceptScore W3040607105C62799726 @default.
- W3040607105 hasConceptScore W3040607105C86803240 @default.
- W3040607105 hasConceptScore W3040607105C99730327 @default.
- W3040607105 hasLocation W30406071051 @default.
- W3040607105 hasOpenAccess W3040607105 @default.
- W3040607105 hasPrimaryLocation W30406071051 @default.
- W3040607105 hasRelatedWork W1537656544 @default.
- W3040607105 hasRelatedWork W1585979398 @default.
- W3040607105 hasRelatedWork W2047467550 @default.
- W3040607105 hasRelatedWork W2070794004 @default.
- W3040607105 hasRelatedWork W2119340451 @default.
- W3040607105 hasRelatedWork W2210352761 @default.
- W3040607105 hasRelatedWork W2256432588 @default.
- W3040607105 hasRelatedWork W2319595298 @default.
- W3040607105 hasRelatedWork W2351982038 @default.
- W3040607105 hasRelatedWork W2950091630 @default.
- W3040607105 hasRelatedWork W2962996570 @default.
- W3040607105 hasRelatedWork W2963618313 @default.
- W3040607105 hasRelatedWork W2988034125 @default.
- W3040607105 hasRelatedWork W3034774142 @default.
- W3040607105 hasRelatedWork W3082525919 @default.
- W3040607105 hasRelatedWork W3090857165 @default.