Lower bounds for artificial neural network approximations: A proof that shallow neural networks fail to overcome the curse of dimensionality.
Philipp GrohsShokhrukh IbragimovArnulf JentzenSarah KoppensteinerPublished in: J. Complex. (2023)
Keyphrases
- artificial neural networks
- neural network
- lower bound
- back propagation
- upper bound
- feed forward
- multilayer perceptron
- branch and bound algorithm
- backpropagation neural networks
- activation function
- recurrent neural networks
- pattern recognition
- neural nets
- np hard
- neural network model
- objective function
- branch and bound
- fuzzy logic
- multi layer perceptron
- radial basis function
- ann models
- upper and lower bounds
- quadratic assignment problem
- question answering
- levenberg marquardt
- genetic algorithm
- lower and upper bounds
- learning rules
- training algorithm
- linear logic
- vc dimension
- worst case
- theorem proving
- theorem prover
- optimal solution
- information extraction
- hidden layer
- multi layer
- genetic algorithm ga
- fault diagnosis
- min sum
- natural language processing