ICML
2008
IEEE
15 years 1 months ago
2008
IEEE
In dimensionality reduction approaches, the data are typically embedded in a Euclidean latent space. However for some data sets this is inappropriate. For example, in human motion...