Gaussian Universality of Linear Classifiers with Random Labels in High-Dimension.
Federica GeraceFlorent KrzakalaBruno LoureiroLudovic StephanLenka ZdeborováPublished in: CoRR (2022)
Keyphrases
- linear classifiers
- high dimension
- hyperplane
- multi class
- real valued
- input space
- feature space
- feature selection
- support vector machine
- high dimensional
- small sample
- generalization error
- high dimensional data
- svm classifier
- training set
- training data
- maximum likelihood
- training examples
- pairwise
- principal components
- multi task
- multi label
- low dimensional
- dimensionality reduction
- image processing
- neural network