Sparse Compositional Metric Learning

TL;DR

This paper introduces a sparse compositional framework for metric learning that combines locally discriminative metrics, leading to improved generalization, scalability, and performance across various datasets.

Contribution

It presents a novel sparse combination approach for metric learning, enabling efficient, flexible, and theoretically justified global, multi-task, and local metrics.

Findings

Outperforms state-of-the-art methods in classification accuracy

Reduces the number of parameters needed for metric estimation

Demonstrates scalability and strong generalization on multiple datasets

Abstract

We propose a new approach for metric learning by framing it as learning a sparse combination of locally discriminative metrics that are inexpensive to generate from the training data. This flexible framework allows us to naturally derive formulations for global, multi-task and local metric learning. The resulting algorithms have several advantages over existing methods in the literature: a much smaller number of parameters to be estimated and a principled way to generalize learned metrics to new testing data points. To analyze the approach theoretically, we derive a generalization bound that justifies the sparse combination. Empirically, we evaluate our algorithms on several datasets against state-of-the-art metric learning methods. The results are consistent with our theoretical findings and demonstrate the superiority of our approach in terms of classification performance and scalability.