Deep Reinforcement Learning and Evolutionary Algorithms

I would like through my research to explore the integration of an on-policy, sample-based DRL algorithm that utilises these advancements to dynamically adapt and optimise within the MAP-Elites framework. This integration aims to enhance the DRL-MAP-Elites pipeline by allowing continuous updates from a

richer, more varied dataset, ensuring that the algorithm not only retains valuable past experiences but also incorporates new information more effectively. Robotics and Reinforcement learning