Deep reinforcement learning for autonomous and continual network intrusion detection

Keyphrases

• Ablation Study 25%
• Acquired Knowledge 25%
• Adaptation Time 25%
• Adaptive Detection 25%
• Adaptive Update 25%
• Benign Behavior 25%
• Catastrophic Forgetting 50%
• Conditional Generative Adversarial Network (cGAN) 25%
• Continual Learning 75%
• Continuous Adaptation 25%
• Cyber Threats 25%
• Deep Reinforcement Learning (deep RL) 100%
• Detection Accuracy 25%
• Emerging Patterns 25%
• Evolving Networks 25%
• Evolving Threats 25%
• Generative Adversarial Networks 50%
• GitHub 25%
• Intrusion Detection Systems Architecture 25%
• Learning Process 25%
• Learning-based Systems 25%
• Machine Learning Based 50%
• Major Challenges 25%
• Network Approach 25%
• Network Intrusion Detection 100%
• Network Intrusion Detection System 100%
• Network Traffic 25%
• New Pattern 25%
• Novel Attacks 25%
• Pattern Detection 25%
• Pattern Labelling 25%
• Pattern Learning 25%
• Rapid Evolution 25%
• Reinforcement Learning 50%
• Reinforcement Learning Agent 25%
• Source Code 25%
• Traditional Network 25%
• Traffic Classification 25%
• Traffic Detection 25%
• Traffic Pattern 75%
• Unseen 50%
• Zero-day Attacks 25%

Computer Science

• Catastrophic Forgetting 66%
• Conditional Generative Adversarial Network 33%
• Continual Learning 100%
• Deep Reinforcement Learning 100%
• Detection Accuracy 33%
• Generative Adversarial Networks 66%
• Learning Agent 33%
• Learning Process 33%
• Learning System 66%
• Machine Learning 66%
• Network Based Intrusion Detection System 33%
• network intrusion detection 100%
• Network Intrusion Detection System 100%
• Network Traffic 33%
• Pattern Detection 33%
• Reinforcement Learning 66%
• Source Coding 33%
• System Architecture 33%
• Traditional Threat 33%
• Traffic Pattern 100%
• Zero-Day Attack 33%