Kernel functions based on triplet comparisons
This work addresses a data representation challenge for machine learning practitioners dealing with relative similarity data, though it appears incremental as it builds on prior embedding methods.
The paper tackles the problem of defining kernel functions from similarity triplets, enabling the use of kernel methods on datasets where only relative comparisons are available, by proposing two approaches that correspond to high-dimensional embeddings.
Given only information in the form of similarity triplets "Object A is more similar to object B than to object C" about a data set, we propose two ways of defining a kernel function on the data set. While previous approaches construct a low-dimensional Euclidean embedding of the data set that reflects the given similarity triplets, we aim at defining kernel functions that correspond to high-dimensional embeddings. These kernel functions can subsequently be used to apply any kernel method to the data set.