Network Anomaly Detection with Quantum Machine Learning

使用量子机器学习进行网络异常检测

• 批准号: 569166-2021
• 负责人: Cherkaoui, SoumayaS
• 金额: $ 12.67万
• 依托单位: École Polytechnique de Montréal
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2022
• 资助国家: 加拿大
• 起止时间: 2022-01-01 至 2023-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=754154
• 关键词: Network; Anomaly; Detection; Quantum; Machine

Cyber-intrusions and cyber-attacks can endanger the security of connected systems and bring irreversible damages to their operation. This is particularly true for critical systems such as train fleets, autonomous vehicles, etc. Intrusion detection systems (IDS) are deployed in conjunction with other security mechanisms such as access control and encryption techniques to further secure systems. Network intrusion detection can be performed by identifying the abnormalities in network monitoring data sets and looking for potential problems. However, building effective systems for network anomaly detection faces many challenges in theory and in practice. The goal of the project is to design and develop a framework for network anomaly detection by using quantum machine learning (QML) algorithms. Quantum machine learning has the potential to revolutionarily transform the capacity to detect network anomalies so as to prevent/defend against cyber intrusions and cyber attacks. The main outcome of this partnership is the materialization of a framework for anomaly detection with QML. This framework has the potential to enable: 1) huge advances in execution time, capacity and learning efficiency and 2) significant improvements in anomaly detection capabilities.

网络入侵和网络攻击可能会危及互联系统的安全，并对其运行带来不可逆转的损害。对于火车车队、自动驾驶车辆等关键系统尤其如此。入侵检测系统 (IDS) 与访问控制和加密技术等其他安全机制结合部署，以进一步保护系统。网络入侵检测可以通过识别网络监控数据集中的异常并寻找潜在问题来进行。然而，构建有效的网络异常检测系统在理论和实践中面临着许多挑战。该项目的目标是使用量子机器学习（QML）算法设计和开发网络异常检测框架。量子机器学习有可能彻底改变检测网络异常的能力，从而预防/防御网络入侵和网络攻击。此次合作的主要成果是实现了 QML 异常检测框架。该框架有潜力实现：1）执行时间、容量和学习效率方面的巨大进步；2）异常检测能力的显着改进。