A new method to instill curiosity in reinforcement learning agents
Several real-world tasks have sparse rewards and this poses challenges for the development of reinforcement learning (RL) algorithms. A solution to this problem is to allow an agent to autonomously create a reward for itself, ...