Neighborhood Watch: Representation Learning with Local-Margin Triplet Loss and Sampling Strategy for K-Nearest-Neighbor Image Classification

TL;DR

This paper introduces a local-margin triplet loss with a novel sampling strategy for improved image classification, especially effective with small datasets and limited data augmentation, backed by theoretical insights and experimental validation.

Contribution

It proposes a new local-margin triplet loss and sampling strategy that enhance nearest-neighbor classification performance with theoretical grounding.

Findings

Outperforms softmax and typical triplet loss on multiple datasets

Effective in small sample scenarios with limited data augmentation

Provides a strong baseline for challenging classification tasks

Abstract

Deep representation learning using triplet network for classification suffers from a lack of theoretical foundation and difficulty in tuning both the network and classifiers for performance. To address the problem, local-margin triplet loss along with local positive and negative mining strategy is proposed with theory on how the strategy integrate nearest-neighbor hyper-parameter with triplet learning to increase subsequent classification performance. Results in experiments with 2 public datasets, MNIST and Cifar-10, and 2 small medical image datasets demonstrate that proposed strategy outperforms end-to-end softmax and typical triplet loss in settings without data augmentation while maintaining utility of transferable feature for related tasks. The method serves as a good performance baseline where end-to-end methods encounter difficulties such as small sample data with limited allowable data augmentation.