DistFromKernel

Distance function obtained from a kernel function.

Formal details (for real valued objects, mixed typed rows in analogy): Let \(k: \mathbb{R}^D \times \mathbb{R}^D\rightarrow \mathbb{R}\) be the pairwise function in kernel, when applied to D-vectors. DistFromKernel(dist) corresponds to the distance function \(d(x, y):= \sqrt{k(x, x) + k(y, y) - 2 \cdot k(x, y)}\).

It should be noted that if \(k\) is positive semi-definite, then \(d\) will be a metric and satisfy the triangle inequality.