152
Voted
DAGM
2011
Springer
14 years 27 days ago
2011
Springer
Many unsupervised learning algorithms make use of kernels that rely on the Euclidean distance between two samples. However, the Euclidean distance is optimal for Gaussian distribut...