Facial Feedback for Reinforcement Learning: A Case Study and Offline Analysis Using the TAMER Framework
This work addresses the problem of sparse human feedback in interactive reinforcement learning for AI agents, offering a novel approach to enhance training efficiency, though it is incremental by building on existing TAMER methods.
The study investigated whether agents can learn from human trainers' facial expressions as evaluative feedback in interactive reinforcement learning, using the TAMER framework in the Infinite Mario benchmark with 561 participants. Results showed that using facial expressions and competition improved feedback prediction accuracies, and simulation experiments confirmed that learning solely from predicted facial feedback is possible, with strong prediction models significantly boosting agent performance.
Interactive reinforcement learning provides a way for agents to learn to solve tasks from evaluative feedback provided by a human user. Previous research showed that humans give copious feedback early in training but very sparsely thereafter. In this article, we investigate the potential of agent learning from trainers' facial expressions via interpreting them as evaluative feedback. To do so, we implemented TAMER which is a popular interactive reinforcement learning method in a reinforcement-learning benchmark problem --- Infinite Mario, and conducted the first large-scale study of TAMER involving 561 participants. With designed CNN-RNN model, our analysis shows that telling trainers to use facial expressions and competition can improve the accuracies for estimating positive and negative feedback using facial expressions. In addition, our results with a simulation experiment show that learning solely from predicted feedback based on facial expressions is possible and using strong/effective prediction models or a regression method, facial responses would significantly improve the performance of agents. Furthermore, our experiment supports previous studies demonstrating the importance of bi-directional feedback and competitive elements in the training interface.