NIPS
2004
14 years 25 days ago
2004
We describe a way of using multiple different types of similarity relationship to learn a low-dimensional embedding of a dataset. Our method chooses different, possibly overlappin...