Leveraging Human Guidance for Deep Reinforcement Learning Tasks
It addresses the problem of inefficient human guidance in reinforcement learning for researchers, but is incremental as it surveys existing methods.
This survey reviews five recent frameworks that incorporate human guidance beyond step-by-step action demonstrations to enhance deep reinforcement learning, aiming to reduce human effort and improve task suitability.
Reinforcement learning agents can learn to solve sequential decision tasks by interacting with the environment. Human knowledge of how to solve these tasks can be incorporated using imitation learning, where the agent learns to imitate human demonstrated decisions. However, human guidance is not limited to the demonstrations. Other types of guidance could be more suitable for certain tasks and require less human effort. This survey provides a high-level overview of five recent learning frameworks that primarily rely on human guidance other than conventional, step-by-step action demonstrations. We review the motivation, assumption, and implementation of each framework. We then discuss possible future research directions.