Training Neural Networks is NP-Hard in Fixed Dimension.
Vincent FroeseChristoph HertrichPublished in: NeurIPS (2023)
Keyphrases
- neural network
- np hard
- training process
- training algorithm
- feedforward neural networks
- scheduling problem
- neural network training
- artificial neural networks
- optimal solution
- backpropagation algorithm
- back propagation
- linear programming
- feed forward neural networks
- special case
- multi layer perceptron
- np complete
- approximation algorithms
- error back propagation
- pattern recognition
- feed forward
- training examples
- worst case
- training set
- genetic algorithm
- neural nets
- multi layer
- computational complexity
- np hardness
- multilayer neural network
- recurrent networks
- neural network structure
- training patterns
- set cover
- radial basis function network
- integer programming
- fault diagnosis
- test set
- fuzzy logic
- lower bound
- machine learning