Batch Decorrelation for Active Metric Learning
This work addresses the challenge of batch annotation in active metric learning for perceptual similarity, which is incremental as it builds on prior active learning strategies.
The paper tackles the problem of training perceptual distance metrics using triplet-based similarity assessments, where standard active learning methods degrade when annotations are requested in batches due to correlation among triplets. The proposed decorrelation method outperforms state-of-the-art approaches, as indicated by experiments.
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.