eGAN: Unsupervised approach to class imbalance using transfer learning
It addresses class imbalance, a common problem in classification tasks, by introducing a novel unsupervised method without synthetic image augmentation, which could benefit practical applications with skewed data.
The paper tackles class imbalance in machine learning by proposing an unsupervised approach using a transfer learning-based encoder GAN (eGAN), achieving a best result of 0.69 F1-score on CIFAR-10 with a 1:2500 imbalance ratio.
Class imbalance is an inherent problem in many machine learning classification tasks. This often leads to trained models that are unusable for any practical purpose. In this study we explore an unsupervised approach to address these imbalances by leveraging transfer learning from pre-trained image classification models to encoder-based Generative Adversarial Network (eGAN). To the best of our knowledge, this is the first work to tackle this problem using GAN without needing to augment with synthesized fake images. In the proposed approach we use the discriminator network to output a negative or positive score. We classify as minority, test samples with negative scores and as majority those with positive scores. Our approach eliminates epistemic uncertainty in model predictions, as the P(minority) + P(majority) need not sum up to 1. The impact of transfer learning and combinations of different pre-trained image classification models at the generator and discriminator is also explored. Best result of 0.69 F1-score was obtained on CIFAR-10 classification task with imbalance ratio of 1:2500. Our approach also provides a mechanism of thresholding the specificity or sensitivity of our machine learning system. Keywords: Class imbalance, Transfer Learning, GAN, nash equilibrium