Testing and learning on distributions with symmetric noise invariance
Kernel embeddings of distributions and the Maximum Mean Discrepancy (MMD), the resulting distance between distributions, are useful tools for fully nonparametric two-sample testing and learning on distributions. However, it is rarely that all possible differences between samples are of interest – di...
| Main Authors: | , , |
|---|---|
| Format: | Conference item |
| Published: |
Curran Associates
2018
|