SelectNet: Learning to Sample from the Wild for Imbalanced Data Training
Yunru Liu, Tingran Gao, Haizhao Yang · 2019
Supervised learning from training data with imbalanced class sizes, a commonly encountered scenario real applications such as anomaly/fraud detection, has long been considered a significant challenge machine learning. Motivated by recent progress curriculum and self-paced learning, we propose to adopt a semi-supervised learning paradigm by training a deep neural network, referred to as SelectNet, to selectively add unlabelled data together with their predicted labels to the training dataset. Unlike existing techniques designed to tackle the difficulty dealing with class imbalanced training data such as resampling, cost-sensitive learning, and margin-based learning, SelectNet provides an end-to-end approach for learning from important unlabelled data in the wild that most likely belong to the under-sampled classes the training data, thus gradually mitigates the imbalance the data used for training the classifier. We demonstrate the efficacy of SelectNet through extensive numerical experiments on standard datasets computer vision.