The necessity of depth for artificial neural networks to approximate certain classes of smooth and bounded functions without the curse of dimensionality.
Lukas GononRobin GraeberArnulf JentzenPublished in: CoRR (2023)
Keyphrases
- artificial neural networks
- neural network
- computational intelligence
- linear combination of basis
- continuous functions
- using artificial neural networks
- depth map
- back propagation
- multilayer perceptron
- depth information
- machine learning
- feed forward neural networks
- high quality
- video sequences
- three dimensional
- least squares
- genetic algorithm
- neural network model
- activation function
- data sets
- database