The Connection Between Approximation, Depth Separation and Learnability in Neural Networks.
Eran MalachGilad YehudaiShai Shalev-ShwartzOhad ShamirPublished in: COLT (2021)
Keyphrases
- neural network
- approximation error
- neural nets
- boolean functions
- approximation methods
- pattern recognition
- fuzzy logic
- back propagation
- multi layer
- depth information
- neural network model
- depth map
- multilayer perceptron
- artificial neural networks
- genetic algorithm
- feed forward
- training process
- dnf formulas
- uniform convergence
- fuzzy systems
- inductive logic programming
- sufficient conditions
- training data
- decision trees
- machine learning