Batch Decorrelation for Active Metric Learning

TL;DR

This paper introduces a novel batch decorrelation method for active metric learning that improves the selection of triplets by balancing informativeness and diversity, leading to better performance in perceptual similarity tasks.

Contribution

The paper proposes a new decorrelation technique for batch active learning in metric learning, addressing correlation issues among triplets and enhancing model training.

Findings

Outperforms state-of-the-art methods in experiments.

Effective decorrelation improves triplet selection quality.

Method is adaptable to various active learning scenarios.

Abstract

We present an active learning strategy for training parametric models of distance metrics, given triplet-based similarity assessments: object $x_i$ is more similar to object $x_j$ than to $x_k$. In contrast to prior work on class-based learning, where the fundamental goal is classification and any implicit or explicit metric is binary, we focus on {\em perceptual} metrics that express the {\em degree} of (dis)similarity between objects. We find that standard active learning approaches degrade when annotations are requested for {\em batches} of triplets at a time: our studies suggest that correlation among triplets is responsible. In this work, we propose a novel method to {\em decorrelate} batches of triplets, that jointly balances informativeness and diversity while decoupling the choice of heuristic for each criterion. Experiments indicate our method is general, adaptable, and outperforms the state-of-the-art.