Selective Unsupervised Feature Learning with Convolutional Neural Network (S-CNN)

TL;DR

This paper introduces S-CNN, a simple and fast unsupervised CNN training method that produces discriminative features for object recognition, reducing reliance on labeled data while achieving competitive accuracy.

Contribution

The paper presents a novel unsupervised training algorithm for CNNs that is easier to train and resistant to overfitting, with competitive performance on standard datasets.

Findings

Achieves accuracy comparable to sophisticated methods

Requires less manual labeling and training effort

Provides discriminative features that generalize well

Abstract

Supervised learning of convolutional neural networks (CNNs) can require very large amounts of labeled data. Labeling thousands or millions of training examples can be extremely time consuming and costly. One direction towards addressing this problem is to create features from unlabeled data. In this paper we propose a new method for training a CNN, with no need for labeled instances. This method for unsupervised feature learning is then successfully applied to a challenging object recognition task. The proposed algorithm is relatively simple, but attains accuracy comparable to that of more sophisticated methods. The proposed method is significantly easier to train, compared to existing CNN methods, making fewer requirements on manually labeled training data. It is also shown to be resistant to overfitting. We provide results on some well-known datasets, namely STL-10, CIFAR-10, and CIFAR-100. The results show that our method provides competitive performance compared with existing alternative methods. Selective Convolutional Neural Network (S-CNN) is a simple and fast algorithm, it introduces a new way to do unsupervised feature learning, and it provides discriminative features which generalize well.