PARD-SSM: Probabilistic Cyber-Attack Regime Detection via Variational Switching State-Space Models

Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• PARD-SSM models network telemetry as a Regime-Dependent Switching Linear Dynamical System with K = 4 hidden regimes corresponding to attack phases.method 95%
• PARD-SSM's structured variational approximation reduces inference complexity from exponential to O(TK²), enabling real-time detection on standard CPU hardware.technical 90%
• PARD-SSM achieves F1 scores of 98.2% on CICIDS2017 dataset and 97.1% on UNSW-NB15 dataset.result 95%
• PARD-SSM has latency less than 1.2 ms per flow for detection.result 90%
• PARD-SSM produces predictive alerts approximately 8 minutes before attack onset, a capability absent in prior systems.result 85%
• Existing signature-based intrusion detection systems cannot detect zero-day attacks.limitation 80%
• Deep-learning intrusion detection models provide opaque anomaly scores without stage attribution.limitation 80%
• Standard Kalman Filters cannot model non-stationary multi-modal dynamics needed for attack phase detection.limitation 80%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card