Quantum-Inspired Algorithms and Perspectives for Optimization
Gerardo Iovane
Abstract
This paper starts with an updated review and analyzes recent developments in quantum-inspired algorithms for cybersecurity, with specific attention to possible perspectives of optimization. The enhancement of classical computing capabilities with quantum principles is transforming fields such as machine learning, optimization, and cybersecurity. Evolutionary algorithms are one example where progress has already been made using quantum techniques through increased efficiency, generalization, and problem-solving techniques exploited by quantum principles. Quantum-inspired evolutionary algorithms (QIEAs) and quantum kernel methods are prime examples of such approaches. Quantum techniques are also used in the field of cybersecurity: QML-based identification systems for intrusion detection strengthen threat detection and encoding through quantum techniques with advanced cryptographic security, while quantum-secure hashing (QSHA) offers sophisticated means of protecting sensitive information. More specifically, QGANs are known for their integration into adversarial generative networks that increase efficiency by replacing classical models in adversarial defense through the generation of synthetic attack models. In this work, a set of benchmarks is provided for comparison with classical and other quantum-inspired technologies. The results demonstrate that these methods far outperform others in terms of computational efficiency and satisfactory scalability. Although fully functional models are still awaited, quantum computing benefits greatly from quantum-inspired technologies, as the latter enable the development of frameworks that bring us closer to the quantum era. Consequently, the work takes the form of an updated systematic review enriched with optimized perspectives.