Reinforcement learning based on human-computer interaction

A novel interactive learning structure integrated with a reinforcement learning algorithm and human-computer interaction (HCI) is proposed. This interactive learning system can benefit from measurements of the distance between the current state and goal state via an operator´s professional knowledge. Thus, the learning procedure is expected to be more efficient. A guess-number task is explored to evaluate the proposed learning system. Experimental results show that the learning efficiency and convergence rate are both increased compared with the normal reinforcement learning method.