Agentic AI for e-governance: a multi-agent reinforcement learning framework for autonomous cyber defence

Key Points

• This research aims to develop a multi-agent reinforcement learning framework for improving cyber defence mechanisms in e-governance systems.
• Designed a multi-agent reinforcement learning framework for cyber defence applications.
• Evaluated the framework in simulated e-governance environments.
• Analyzed the security response performance of the framework.
• Demonstrated improved incident response times in autonomous cyber defence.
• Achieved a reduction in vulnerability metrics by 25% over baseline simulations.
• Showed a significant increase in threat detection accuracy, reaching 90%.