A Quantum LSTM Based Approach to Cyber Threat Detection in Virtual Environment

Abstract

Abstract Quantum computing is the application of quantum phenomena, such as superposition and entanglement, to perform probabilistic computations in the area of information processing. Quantum Information Processing (QIP) holds the promise of having a significant speed advantage over classical processing. This advantage very naturally extends to the advancements in Artificial Intelligence / Machine learning (AI/ML) in the form of what is now conveniently referred to as Quantum Artificial Intelligence (QAI) / Quantum Machine Learning (QML). Traditional AI/ML algorithms are designed to efficiently identify patterns from datasets, and consequently, there is a huge body of work in classical AI/ML on anomaly detection techniques. This work has also been successfully applied in the area of Cyber Security. Automated, advanced methods of attack vector recognition using virtual machine introspection have been successfully studied via Classical AI/ML algorithms such as Long Short-Term Memory (LSTM). These methods, however, face the large dataset handling and real-time processing limitations of classical machines, and one is therefore presented with the same question: Can QML provide an advantage in this scenario? This is what we will explore in this paper.