Matches in SemOpenAlex for { <https://semopenalex.org/work/W1880557952> ?p ?o ?g. }
Showing items 1 to 58 of
58
with 100 items per page.
- W1880557952 abstract "The main purpose of this talk is to introduce auseful tool for the analysis of discrete neural networksin which every node is a Boolean thresholdgate. The difficulty in the analysis of neuralnetworks arises from the fact that the basicprocessing elements (linear threshold gates) arenonlinear. The key idea in harmonic analysisof threshold functions is to represent the functionsas polynomials over the field of real numbers.Answering different questions regardingneural networks becomes equivalent to answeringquestions related to the coefficients of thesepolynomials. We have applied these techniquesand obtained many interesting and surprising results[1, 2, 3, 4]. The focus of this talk willbe on presenting a theorem that characterizes-usingspetral norms-the complexity of computinga Boolean function with threshold circuits[2, 3]. This result establishes the first known linkbetween harmonic analysis and the complexity ofcomputing with neural networks." @default.
- W1880557952 created "2016-06-24" @default.
- W1880557952 creator A5043861677 @default.
- W1880557952 creator A5044935694 @default.
- W1880557952 date "2005-08-24" @default.
- W1880557952 modified "2023-09-24" @default.
- W1880557952 title "Harmonic Analysis And The Complexity Of Computing With Threshold (Neural) Elements" @default.
- W1880557952 cites W2097575812 @default.
- W1880557952 cites W2117508909 @default.
- W1880557952 doi "https://doi.org/10.1109/isit.1991.695142" @default.
- W1880557952 hasPublicationYear "2005" @default.
- W1880557952 type Work @default.
- W1880557952 sameAs 1880557952 @default.
- W1880557952 citedByCount "0" @default.
- W1880557952 crossrefType "proceedings-article" @default.
- W1880557952 hasAuthorship W1880557952A5043861677 @default.
- W1880557952 hasAuthorship W1880557952A5044935694 @default.
- W1880557952 hasBestOaLocation W18805579522 @default.
- W1880557952 hasConcept C11413529 @default.
- W1880557952 hasConcept C121332964 @default.
- W1880557952 hasConcept C127413603 @default.
- W1880557952 hasConcept C127934551 @default.
- W1880557952 hasConcept C131770355 @default.
- W1880557952 hasConcept C154945302 @default.
- W1880557952 hasConcept C179799912 @default.
- W1880557952 hasConcept C24326235 @default.
- W1880557952 hasConcept C24890656 @default.
- W1880557952 hasConcept C41008148 @default.
- W1880557952 hasConcept C50644808 @default.
- W1880557952 hasConceptScore W1880557952C11413529 @default.
- W1880557952 hasConceptScore W1880557952C121332964 @default.
- W1880557952 hasConceptScore W1880557952C127413603 @default.
- W1880557952 hasConceptScore W1880557952C127934551 @default.
- W1880557952 hasConceptScore W1880557952C131770355 @default.
- W1880557952 hasConceptScore W1880557952C154945302 @default.
- W1880557952 hasConceptScore W1880557952C179799912 @default.
- W1880557952 hasConceptScore W1880557952C24326235 @default.
- W1880557952 hasConceptScore W1880557952C24890656 @default.
- W1880557952 hasConceptScore W1880557952C41008148 @default.
- W1880557952 hasConceptScore W1880557952C50644808 @default.
- W1880557952 hasLocation W18805579521 @default.
- W1880557952 hasLocation W18805579522 @default.
- W1880557952 hasOpenAccess W1880557952 @default.
- W1880557952 hasPrimaryLocation W18805579521 @default.
- W1880557952 hasRelatedWork W1554521983 @default.
- W1880557952 hasRelatedWork W170887098 @default.
- W1880557952 hasRelatedWork W2011473608 @default.
- W1880557952 hasRelatedWork W2046614375 @default.
- W1880557952 hasRelatedWork W2073676219 @default.
- W1880557952 hasRelatedWork W2121410378 @default.
- W1880557952 hasRelatedWork W2168287377 @default.
- W1880557952 hasRelatedWork W2169141426 @default.
- W1880557952 hasRelatedWork W2369653905 @default.
- W1880557952 hasRelatedWork W2512937901 @default.
- W1880557952 isParatext "false" @default.
- W1880557952 isRetracted "false" @default.
- W1880557952 magId "1880557952" @default.
- W1880557952 workType "article" @default.