Advancing anomaly detection in cloud environments with cutting-edge generative AI for expert systems

Abstract

As artificial intelligence (AI) continues to advance, Generative AI emerges as a transformative force, capable of generating novel content and revolutionizing anomaly detection methodologies. This paper presents CloudGEN, a pioneering approach to anomaly detection in cloud environments by leveraging the potential of Generative Adversarial Networks (GANs) and Convolutional Neural Network (CNN). Our research focuses on developing a state-of-the-art Generative AI-based anomaly detection system, integrating GANs, deep learning techniques, and adversarial training. We explore unsupervised generative modelling, multi-modal architectures, and transfer learning to enhance expert systems' anomaly detection systems. We illustrate our approach by dissecting anomalies regarding job performance, network behaviour, and resource utilization in cloud computing environments. The experimental results underscore a notable surge in anomaly detection accuracy with significant development of approximately 11%.