116
Voted
DAGM
2011
Springer
13 years 9 months 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...