From subspace learning to distance learning: a geometrical optimization approach

In this paper, we adopt a differential-geometry viewpoint to tackle the problem of learning a distance online. As this prob- lem can be cast into the estimation of a fixed-rank positive semidefinite (PSD) matrix, we develop algorithms that ex- ploits the rich geometry structure of the set of fixed-rank PSD matrices. We propose a method which separately updates the subspace of the matrix and its projection onto that subspace. A proper weighting of the two iterations enables to continu- ously interpolate between the problem of learning a subspace and learning a distance when the subspace is fixed.