paper
Learning Red Agent Policy from Observations for Neurosymbolic Autonomous Cyber Agents
paperactiveprovisional
learning-red-agent-policy-from-observations-for-neurosymbolic-autonomous-cyber-agents-33782979·1 events·first seen 9h agoAliases: Learning Red Agent Policy from Observations for Neurosymbolic Autonomous Cyber Agents
Co-occurring entities
More like this (12)
Agentopia: Long-Term Life Simulation and Learning in Agent SocietiesGeometric Action Model for Robot Policy LearningTowards a Science of AI Agent ReliabilityPreference Coordinated Multi-agent Policy OptimizationReward Modeling for Multi-Agent OrchestrationAPPO: Agentic Procedural Policy OptimizationVisual Verification Enables Inference-time Steering and Autonomous Policy ImprovementCollaborative Human-Agent ProtocolAn Agency-Transferring Model-Free Policy Enhancement TechniqueA History-Aware Visually Grounded Critic for Computer Use AgentsHow AI Agents Reshape Knowledge Work: Autonomy, Efficiency, and ScopeSemantic Agent
Recent events (1)
Imitation learning technique infers red agent policy in partially observable cyber-defense environments
Researchers propose a Policy Learning Technique using imitation learning to infer attacker (red agent) policies from network observations and defender actions in partially observable autonomous cyber environments. The method integrates with neurosymbolic cyber-defense agents that use behavior trees with learning-enabled components. Evaluated across diverse simulated scenarios, the approach achieves high prediction accuracy for red agent actions, improving the defender's ability to anticipate intrusions.