Universal Approximation Theorem for Vector- and Hypercomplex-Valued Neural Networks.
Marcos Eduardo ValleWington L. VitalGuilherme VieiraPublished in: CoRR (2024)
Keyphrases
- neural network
- euclidean norm
- approximation error
- multi valued
- pattern recognition
- fourier transform
- error bounds
- closed form
- fuzzy logic
- artificial neural networks
- sparse matrix
- recurrent neural networks
- approximation algorithms
- fuzzy systems
- taylor series
- back propagation
- neural nets
- genetic algorithm
- valued logic
- vector data
- von neumann
- image processing
- queueing networks
- feed forward
- feature vectors
- color images
- image features
- vector space
- image representation
- image classification