L1-regularized Neural Networks are Improperly Learnable in Polynomial Time.
Yuchen ZhangJason D. LeeMichael I. JordanPublished in: ICML (2016)
Keyphrases
- neural network
- dnf formulas
- exact learning
- pattern recognition
- fuzzy logic
- equivalence queries
- equivalence and membership queries
- dnf formulae
- recurrent neural networks
- special case
- computational complexity
- neural network model
- artificial neural networks
- membership queries
- genetic algorithm
- back propagation
- training data
- multilayer perceptron
- self organizing maps
- multi layer
- positive examples
- least squares
- positive data
- feed forward
- total least squares
- fault diagnosis
- neural nets
- hypothesis space
- concept classes
- statistical queries
- risk minimization
- loss function