Improving unsupervised learning with exemplar CNNs
Publication Type:
Refereed Conference Meeting Proceeding
Abstract:
Most recent unsupervised learning methods explore alternative objectives, often referred to as self-supervised tasks, to train convolutional neural networks without the supervision of human annotated labels. This paper explores the generation of surrogate classes as a self-supervised alternative to learn discriminative features, and proposes a clustering algorithm to overcome one of the main limitations of this kind of approach. Our clustering technique improves the initial implementation and achieves 76.4% accuracy in the STL-10 test set, surpassing the current state-ofthe-art for the STL-10 unsupervised benchmark. We also explore several issues with the unlabeled set from STL-10 that should be considered in future research using this dataset.
Conference Name:
Irish Machine Vision and Image Processing conference
Digital Object Identifer (DOI):
10.21427/9mp9-0t26
Publication Date:
28/08/2019
Conference Location:
Ireland
Research Group:
Institution:
Dublin City University (DCU)
Open access repository:
Yes