Data Augmentation Using GANs

TL;DR

This paper explores using GANs to generate synthetic training data to improve machine learning performance, especially in cases of imbalanced or sensitive datasets, demonstrating comparable or better results than original data.

Contribution

It introduces a novel application of GANs for data augmentation in classification tasks, showing effectiveness across various datasets and architectures.

Findings

GAN-generated data achieves similar or better accuracy than original data

Effective in imbalanced and sensitive data scenarios

Decision Tree classifiers perform well with GAN-augmented data

Abstract

In this paper we propose the use of Generative Adversarial Networks (GAN) to generate artificial training data for machine learning tasks. The generation of artificial training data can be extremely useful in situations such as imbalanced data sets, performing a role similar to SMOTE or ADASYN. It is also useful when the data contains sensitive information, and it is desirable to avoid using the original data set as much as possible (example: medical data). We test our proposal on benchmark data sets using different network architectures, and show that a Decision Tree (DT) classifier trained using the training data generated by the GAN reached the same, (and surprisingly sometimes better), accuracy and recall than a DT trained on the original data set.