Quantum Machine Learning for Cyber Threat Detection: Enhancing Security in the Quantum Era

The cat-and-mouse game is endless. On the one hand, the security industry is ceaselessly countering security technologies to immunize against perceptible threats. On the other hand, adversaries are always attempting to devise cutting-edge exfiltration schemes for potential return on investment. Ever since the Federal Bureau of Investigation (FBI) released the Flash report that hinges on raw indicators to unveil LICAT backdoor on Joomla websites running PHP, adversaries tweaked their infection vectors to avoid detectable host-based artifacts. Economic and geopolitical aspects also spurred the entrenchment of Advanced Persistent Threats (APTs). Major players in cyber industry are investing manifold resources into protracted espionage campaigns by concentrating on high-value targets. While research in QC is progressing, there are also challenges in harnessing the technology’s full capabilities. Much like conventional computing requiring some understanding about a computer’s architecture, a fundamental understanding of quantum mechanics is needed to comprehend QC . To explore how advances in quantum may be applied in cybersecurity tasks, developments in this technology are reviewed. Quantum computing can help to deal with cybersecurity tasks by enhancing problem complexity due to its advantages in performing complex tasks effectively. Cybersecurity tasks may be substantially complex, requiring understanding non-local correlations and collecting a large quantity of data points in different fidelities and complexities.