Semi-Supervised Deep Learning Using Improved Unsupervised Discriminant Projection

TL;DR

This paper introduces a semi-supervised deep learning method that leverages an improved unsupervised discriminant projection as a regularization to effectively utilize unlabeled data, reducing the need for extensive labeled datasets.

Contribution

It proposes a novel semi-supervised deep learning algorithm that incorporates an improved discriminant projection to utilize unlabeled data for better classification accuracy.

Findings

Achieves satisfactory classification with only dozens of labeled samples.

Effectively utilizes local and nonlocal distributions of unlabeled data.

Improves deep learning performance in data-scarce scenarios.

Abstract

Deep learning demands a huge amount of well-labeled data to train the network parameters. How to use the least amount of labeled data to obtain the desired classification accuracy is of great practical significance, because for many real-world applications (such as medical diagnosis), it is difficult to obtain so many labeled samples. In this paper, modify the unsupervised discriminant projection algorithm from dimension reduction and apply it as a regularization term to propose a new semi-supervised deep learning algorithm, which is able to utilize both the local and nonlocal distribution of abundant unlabeled samples to improve classification performance. Experiments show that given dozens of labeled samples, the proposed algorithm can train a deep network to attain satisfactory classification results.