Iterative Self-Learning: Semi-Supervised Improvement to Dataset Volumes and Model Accuracy

Robert Dupre, Jiri Fajtl, Vasileios Argyriou, Paolo Remagnino; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2019, pp. 79-82

Abstract


A novel semi-supervised learning technique is introduced based on a simple iterative learning cycle together with learned thresholding techniques and an ensemble decision support system. State-of-the-art model performance and increased training data volume are demonstrated, through the use of unlabelled data when training deeply learned classification models. Evaluation of the proposed approach is performed on commonly used datasets when evaluating semi-supervised learning techniques as well as a number of more challenging image classification datasets (CIFAR-100 and a 200 class subset of ImageNet).

Related Material


[pdf]
[bibtex]
@InProceedings{Dupre_2019_CVPR_Workshops,
author = {Dupre, Robert and Fajtl, Jiri and Argyriou, Vasileios and Remagnino, Paolo},
title = {Iterative Self-Learning: Semi-Supervised Improvement to Dataset Volumes and Model Accuracy},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2019}
}