Matches in SemOpenAlex for { <https://semopenalex.org/work/W4280524871> ?p ?o ?g. }
Showing items 1 to 59 of
59
with 100 items per page.
- W4280524871 abstract "Since statistical guarantees for neural networks are usually restricted to global optima of intricate objective functions, it is not clear whether these theories really explain the performances of actual outputs of neural-network pipelines. The goal of this paper is, therefore, to bring statistical theory closer to practice. We develop statistical guarantees for simple neural networks that coincide up to logarithmic factors with the global optima but apply to stationary points and the points nearby. These results support the common notion that neural networks do not necessarily need to be optimized globally from a mathematical perspective. More generally, despite being limited to simple neural networks for now, our theories make a step forward in describing the practical properties of neural networks in mathematical terms." @default.
- W4280524871 created "2022-05-22" @default.
- W4280524871 creator A5019872107 @default.
- W4280524871 creator A5030418930 @default.
- W4280524871 creator A5091328279 @default.
- W4280524871 date "2022-05-09" @default.
- W4280524871 modified "2023-09-23" @default.
- W4280524871 title "Statistical Guarantees for Approximate Stationary Points of Simple Neural Networks" @default.
- W4280524871 doi "https://doi.org/10.48550/arxiv.2205.04491" @default.
- W4280524871 hasPublicationYear "2022" @default.
- W4280524871 type Work @default.
- W4280524871 citedByCount "0" @default.
- W4280524871 crossrefType "posted-content" @default.
- W4280524871 hasAuthorship W4280524871A5019872107 @default.
- W4280524871 hasAuthorship W4280524871A5030418930 @default.
- W4280524871 hasAuthorship W4280524871A5091328279 @default.
- W4280524871 hasBestOaLocation W42805248711 @default.
- W4280524871 hasConcept C111472728 @default.
- W4280524871 hasConcept C126255220 @default.
- W4280524871 hasConcept C12713177 @default.
- W4280524871 hasConcept C134306372 @default.
- W4280524871 hasConcept C138885662 @default.
- W4280524871 hasConcept C147168706 @default.
- W4280524871 hasConcept C154945302 @default.
- W4280524871 hasConcept C2780586882 @default.
- W4280524871 hasConcept C33923547 @default.
- W4280524871 hasConcept C39927690 @default.
- W4280524871 hasConcept C41008148 @default.
- W4280524871 hasConcept C50644808 @default.
- W4280524871 hasConcept C86582703 @default.
- W4280524871 hasConceptScore W4280524871C111472728 @default.
- W4280524871 hasConceptScore W4280524871C126255220 @default.
- W4280524871 hasConceptScore W4280524871C12713177 @default.
- W4280524871 hasConceptScore W4280524871C134306372 @default.
- W4280524871 hasConceptScore W4280524871C138885662 @default.
- W4280524871 hasConceptScore W4280524871C147168706 @default.
- W4280524871 hasConceptScore W4280524871C154945302 @default.
- W4280524871 hasConceptScore W4280524871C2780586882 @default.
- W4280524871 hasConceptScore W4280524871C33923547 @default.
- W4280524871 hasConceptScore W4280524871C39927690 @default.
- W4280524871 hasConceptScore W4280524871C41008148 @default.
- W4280524871 hasConceptScore W4280524871C50644808 @default.
- W4280524871 hasConceptScore W4280524871C86582703 @default.
- W4280524871 hasLocation W42805248711 @default.
- W4280524871 hasOpenAccess W4280524871 @default.
- W4280524871 hasPrimaryLocation W42805248711 @default.
- W4280524871 hasRelatedWork W2038803387 @default.
- W4280524871 hasRelatedWork W2148454247 @default.
- W4280524871 hasRelatedWork W2368259199 @default.
- W4280524871 hasRelatedWork W2386387936 @default.
- W4280524871 hasRelatedWork W2627078154 @default.
- W4280524871 hasRelatedWork W3169924058 @default.
- W4280524871 hasRelatedWork W4238075012 @default.
- W4280524871 hasRelatedWork W4248389398 @default.
- W4280524871 hasRelatedWork W4280524871 @default.
- W4280524871 hasRelatedWork W2247596074 @default.
- W4280524871 isParatext "false" @default.
- W4280524871 isRetracted "false" @default.
- W4280524871 workType "article" @default.