Reinforcement-Learning-based-Self-Healing-Logistics-Databases-for-Disaster-Response-Operations

To create a custom environment that simulates the behavior of a database under stress and allows an RL agent to learn how to take corrective actions to maintain uptime and data quality.

Objectives for Real-Time Evaluation

• Deploy the trained DQN agent on a live PostgreSQL stream simulation.
• Evaluate its actions using realistic anomalies.
• Collect and log meaningful performance metrics.