A Decoder-Free Approach for Unsupervised Clustering and Manifold Learning with Random Triplet Mining

Oliver Nina, Jamison Moody, Clarissa Milligan; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 0-0

Abstract


Unsupervised clustering is a very relevant open area of research in machine learning with many applications in the real world. Learning the manifold in which images lie and measuring the proximity distance of the sample points to the clusters in their latent space is non-trivial. Recent deep learning methods have proposed the use of autoencoders for manifold learning and dimensionality reduction in an effort to better cluster image samples. However, offline training of autoencoders is cumbersome and rather tedious to update. Moreover, trained autoencoders tend to be biased towards the training set and are impractical for performing data augmentation. In this paper, we introduce a novel method that uses a triplet network architecture in order to replace autoencoders, thus avoiding the need to pre-train autoencoders offline. Because our framework can be trained online, we can train our network with data augmented pairs which allows us to build a more robust encoder and improve accuracy. In contrast to other clustering methods that require nearest neighbor comparisons at every step, our method introduces a novel and adaptive approach of choosing the samples to train which we call Random Triplet Mining. Our method remains competitive compared with other current methods while we obtain state of the art results on the Fashion-MNIST dataset.

Related Material


[pdf]
[bibtex]
@InProceedings{Nina_2019_ICCV,
author = {Nina, Oliver and Moody, Jamison and Milligan, Clarissa},
title = {A Decoder-Free Approach for Unsupervised Clustering and Manifold Learning with Random Triplet Mining},
booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops},
month = {Oct},
year = {2019}
}