Multiverse Computing and CounterCraft researchers have developed a quantum AI model that significantly improves cyber attack detection. The Matrix Product State (MPS) model, trained on real network traffic and system logs, identified 100% of attacks. The model reduces false positives and improves the explainability of results, an important feature for businesses and regulators.
The model also creates synthetic data for training and simulating activity. The software includes a risk tolerance slider for security analysts to adjust threat detection sensitivity. This work could enhance anomaly detection across various sectors, including finance, healthcare, and government.
Quantum-Inspired Algorithm Enhances Cyber Attack Detection
Researchers from Multiverse Computing and CounterCraft, both companies specializing in quantum computing solutions and threat intelligence respectively, have developed a new quantum AI model that significantly enhances the detection of cyber attacks. The Matrix Product State (MPS) model, trained on datasets from actual network traffic and system logs, was able to identify all attacks in the test data.
The MPS model uses threat intelligence generated by adversaries, a departure from traditional rule-based systems, to identify cyber attacks. This approach offers improved interpretability and clear insights into anomalies identified by the algorithm. The researchers believe that continuous advancements and enhancements to the model’s interpretability capabilities will facilitate its real-world application in the near future.
MPS Model: A Benchmark for Cybersecurity
The performance of the MPS model was evaluated against a benchmark provided by CounterCraft analysts who had already identified the attacks within the training data. The model’s effectiveness is detailed in a new paper submitted to arXiv: “Tensor Networks for Explainable Machine Learning in Cybersecurity.”
Compared to most classical models, the MPS model excels in reducing false positives with greater precision. It also improves the explainability of the algorithm’s results, a capability that is becoming increasingly important for business users and regulators.
The Importance of Explainable AI in Cybersecurity
Explainable AI, which provides clear explanations for outcomes, supports robust decision-making, improves understanding of threats, and ensures compliance with increasingly stringent transparency regulations. The collaboration between Multiverse Computing and CounterCraft demonstrates how quantum techniques can enhance cybersecurity defenses against current and future threats while improving explainability.
A cyber attack typically consists of a series of 20-80 individual events. The MPS model was able to identify 83.5% of these steps and even detected several steps missed in the classical analysis.
Training Data and Detection Capabilities
The training data for the MPS model, provided by CounterCraft, contained detailed incident reports covering various attack types, such as weak credential usage and exploits of known vulnerabilities. The model was trained to distinguish between normal and abnormal behavior, enabling it to identify attacks.
The model also generates synthetic data, which can be used for training models and simulating activity for deception strategies. It includes a user-friendly risk tolerance slider, allowing security analysts to adjust the sensitivity of the threat detection to their requirements. This ensures a high detection rate with a manageable number of alerts.
Future Research and Applications
This research lays the groundwork for future studies in cybersecurity and quantum software. Potential next steps could include robust testing to enhance the model’s effectiveness in diverse scenarios. While the initial use case was cybersecurity, this model has the potential to enhance anomaly detection across various sectors, including finance, healthcare, government, critical infrastructure, manufacturing, and retail.
External Link: Click Here For More
