Transferable Deep Metric Learning for Clustering

TL;DR

This paper introduces a transferable deep metric learning framework that learns a single metric on one dataset and applies it to cluster different datasets effectively, addressing high-dimensional clustering challenges.

Contribution

It presents a novel method for learning a universal metric transferable across datasets, reducing the need for dataset-specific metric learning.

Findings

Achieves competitive clustering results on multiple datasets

Uses only a few labeled datasets and shallow networks

Demonstrates effectiveness across synthetic and real-world data

Abstract

Clustering in high dimension spaces is a difficult task; the usual distance metrics may no longer be appropriate under the curse of dimensionality. Indeed, the choice of the metric is crucial, and it is highly dependent on the dataset characteristics. However a single metric could be used to correctly perform clustering on multiple datasets of different domains. We propose to do so, providing a framework for learning a transferable metric. We show that we can learn a metric on a labelled dataset, then apply it to cluster a different dataset, using an embedding space that characterises a desired clustering in the generic sense. We learn and test such metrics on several datasets of variable complexity (synthetic, MNIST, SVHN, omniglot) and achieve results competitive with the state-of-the-art while using only a small number of labelled training datasets and shallow networks.