Machine learning with limited data

TL;DR

This paper addresses the challenge of few-shot image classification by proposing novel data augmentation and attention methods, and extends to cross-domain scenarios with unlabeled data, achieving significant improvements over baselines.

Contribution

It introduces two new methods for few-shot learning, including style augmentation and spatial attention, and proposes a framework for cross-domain few-shot learning with unlabeled data.

Findings

Proposed style augmentation improves classification accuracy.

Spatial attention enhances relations between image patches.

Methods outperform baseline models significantly.

Abstract

Thanks to the availability of powerful computing resources, big data and deep learning algorithms, we have made great progress on computer vision in the last few years. Computer vision systems begin to surpass humans in some tasks, such as object recognition, object detection, face recognition and pose estimation. Lots of computer vision algorithms have been deployed to real world applications and started to improve our life quality. However, big data and labels are not always available. Sometimes we only have very limited labeled data, such as medical images which requires experts to label them. In this paper, we study few shot image classification, in which we only have very few labeled data. Machine learning with little data is a big challenge. To tackle this challenge, we propose two methods and test their effectiveness thoroughly. One method is to augment image features by mixing the style of these images. The second method is applying spatial attention to explore the relations between patches of images. We also find that domain shift is a critical issue in few shot learning when the training domain and testing domain are different. So we propose a more realistic cross-domain few-shot learning with unlabeled data setting, in which some unlabeled data is available in the target domain. We propose two methods in this setting. Our first method transfers the style information of the unlabeled target dataset to the samples in the source dataset and trains a model with stylized images and original images. Our second method proposes a unified framework to fully utilize all the data. Both of our methods surpass the baseline method by a large margin.