Uniform Convergence of Interpolators: Gaussian Width, Norm Bounds and Benign Overfitting.
Frederic KoehlerLijia ZhouDanica J. SutherlandNathan SrebroPublished in: NeurIPS (2021)
Keyphrases
- uniform convergence
- covering numbers
- generalization error
- gaussian kernels
- empirical risk minimization
- sufficient conditions
- learning rate
- vc dimension
- reproducing kernel hilbert space
- generalization bounds
- risk minimization
- statistical learning theory
- upper and lower bounds
- sample complexity
- cross validation
- gaussian kernel
- upper bound
- model selection
- maximum likelihood
- decision trees
- objective function
- real valued
- learning algorithm
- ranking functions
- distribution free
- large deviations
- training data
- special case
- data dependent
- learning problems
- gaussian mixture model
- kernel function
- kernel machines
- linear classifiers
- statistical learning
- supervised learning
- support vector machine