Spectrum Dependent Learning Curves in Kernel Regression and Wide Neural Networks.
Blake BordelonAbdulkadir CanatarCengiz PehlevanPublished in: ICML (2020)
Keyphrases
- kernel regression
- learning curves
- neural network
- learning curve
- average case
- regression model
- nadaraya watson
- gaussian processes
- artificial neural networks
- parameter optimization
- random graphs
- power law
- vc dimension
- regression methods
- linear regression
- fuzzy logic
- learning algorithm
- sample complexity
- worst case
- genetic algorithm