Benign Overfitting in Deep Neural Networks under Lazy Training.
Zhenyu ZhuFanghui LiuGrigorios ChrysosFrancesco LocatelloVolkan CevherPublished in: ICML (2023)
Keyphrases
- neural network
- training process
- training algorithm
- feedforward neural networks
- backpropagation algorithm
- deep architectures
- feed forward neural networks
- multi layer perceptron
- back propagation
- neural network model
- training set
- fuzzy logic
- radial basis function network
- neural network structure
- genetic algorithm
- artificial neural networks
- breast cancer
- training patterns
- pattern recognition
- lazy learning
- data sets
- recurrent networks
- neural network training
- avoid overfitting
- self organizing maps
- cross validation
- online learning
- decision trees
- training phase
- rule extraction
- fault diagnosis
- test set
- sufficient conditions
- supervised learning
- support vector machine
- expert systems
- bayesian networks
- learning algorithm