Classifying high-dimensional Gaussian mixtures: Where kernel methods fail and neural networks succeed.
Maria RefinettiSebastian GoldtFlorent KrzakalaLenka ZdeborováPublished in: ICML (2021)
Keyphrases
- kernel methods
- gaussian mixture
- high dimensional
- neural network
- feature space
- kernel function
- high dimensional feature space
- gaussian mixture model
- em algorithm
- support vector
- closed form
- probability density function
- expectation maximization
- kernel principal component analysis
- dimensionality reduction
- covariance matrix
- low dimensional
- kernel matrix
- support vector machine
- reproducing kernel hilbert space
- pattern recognition
- kernel pca
- machine learning
- mixture model
- density estimation
- similarity search
- input space
- nearest neighbor
- data points
- high dimensional data
- principal component analysis
- feature vectors
- training data
- feature extraction