Abstract
Advances in cybersecurity are crucial for a country's economic and national security. As data transmission and storage exponentially increase, new threat detection and mitigation techniques are urgently needed. Cybersecurity has become an absolute necessity, with the ever-increasing transmitted networks from day to day causing exponential growth of data that is being stored on servers. In order to thwart sophisticated attacks in the future, it will be necessary to regularly update threat detection and data preservation techniques. Generative adversarial networks (GANs) are a class of unsupervised machine learning models that can generate synthetic data. GANs are gaining importance in AI-based cybersecurity systems for applications such as intrusion detection, steganography, cryptography, and anomaly detection. This paper provides a comprehensive review of research on applying GANs for cybersecurity, including an analysis of popular cybersecurity datasets and GAN model architectures used in these studies.