Intelligent Threat Detection in the Post-Quantum Era
summary
Claims that signature systems miss 38% detection failure and that the assembled ML stack hits a 93% detection rate using random forests, gradient boosting, SVMs, isolation forest, DBSCAN, LSTM and transformer models. Feature selection reportedly boosts model performance by 40% while real‑time pipeline combines vectorized features, risk scoring and automated mitigation in shadow→CI/CD rollout phases. Privacy uses federated learning, differential privacy and homomorphic encryption; post‑quantum readiness folds postquantum crypto standards and crypto agility into model pipelines.
AI-written summary · about
chapters (11)
- 0:00Introduction and Session Overview
- 1:10Evolving Cyber Threat Landscape
- 1:58ML Techniques for Cybersecurity
- 3:35Multimodal Detection Architecture
- 5:11Feature Engineering for Cybersecurity ML
- 6:17Real-Time Classification Pipeline
- 7:00Privacy-Preserving ML Techniques
- 8:28Phased Approach to ML Implementation
- 9:48Operational Excellence and Monitoring
- 10:22Preparing for Post-Quantum Cybersecurity
- 11:45Key Takeaways and Conclusion
related talks
17mTrust at Scale: Fueling AI and Defending Against Quantum Threats
01h 04mCybersecurity in the Era of AI and Quantum Computing
30mPreparing Vulnerability Management for the Post-Quantum Era: From Legacy Cryptography Crypto-Agility
31mBuild threat detection models using ML
42mPrivacy Challenges in Machine Learning
32mAI/Machine Learning: What is Actually Working in Cybersecurity?
19mTransform CyberSec Governance with GenAI
47mQuantum Machine Learning: Cutting Through the Buzzword BS