2019 International Joint Conference on Neural Networks (IJCNN) https://ieeexplore.ieee.org/document/8852242/

Many techniques try to increase the amount of minority class samples. The simplest way is to repeat samples randomly. Other techniques try to use some criterion to select the samples, such as a level of difficulty that they have according to the classification model. Some more elaborate techniques to generate extra samples trying to follow the original distribution of the data, however, much class noise is inserted in the dataset since samples may follow an incorrect assumption of that distribution. To solve this problem, we propose a generative adversarial model architecture using InfoGAN to try to learn the data distributions and generate synthetic samples of ECG beats. The idea is to balance the dataset with sintentic data generated by an InfoGAN. We call this method of Adversarial Oversampling.

We should thank to @eriklindernoren for by your repository of GANs implementations using Keras

We are using samples from MIT-BIH Arrhythmia Database

• model.py- CNN classification model and oversampling evaluation.
• gan_model.py- Training of GAN (InfoGAN) and generation of synthetic samples.