Statements (19)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:mathematical_concept
|
| gptkbp:activatedBy |
sigmoid
|
| gptkbp:alsoKnownAs |
gptkb:Universal_Approximation_Theorem
|
| gptkbp:appliesTo |
feedforward neural networks
single hidden layer networks |
| gptkbp:author |
gptkb:George_Cybenko
|
| gptkbp:citation |
highly cited
|
| gptkbp:field |
gptkb:machine_learning
gptkb:mathematics neural networks |
| gptkbp:impact |
foundation of neural network theory
|
| gptkbp:journalPublished |
gptkb:Mathematics_of_Control,_Signals,_and_Systems
|
| gptkbp:publicationYear |
1989
|
| gptkbp:relatedTo |
gptkb:Hornik's_theorem
gptkb:Kolmogorov–Arnold_representation_theorem |
| gptkbp:sentence |
A feedforward neural network with a single hidden layer containing a finite number of neurons can approximate any continuous function on compact subsets of R^n, given suitable activation functions.
|
| gptkbp:bfsParent |
gptkb:George_Cybenko
|
| gptkbp:bfsLayer |
6
|
| https://www.w3.org/2000/01/rdf-schema#label |
Cybenko theorem
|