Disentangling the Predictive Variance of Deep Ensembles through the Neural Tangent Kernel.
Seijin KobayashiPau Vilimelis AceitunoJohannes von OswaldPublished in: NeurIPS (2022)
Keyphrases
- network architecture
- support vector
- ensemble methods
- neural network
- kernel function
- decision trees
- positive definite
- kernel matrix
- random forests
- ensemble learning
- associative memory
- kernel methods
- bio inspired
- component analysis
- correlation coefficient
- covariance matrix
- weighted voting
- standard deviation
- base classifiers
- similarity function
- feature space