Newtonian Action Advice: Integrating Human Verbal Instruction with Reinforcement Learning
This addresses the challenge of improving human-agent interaction for non-expert users, though it is incremental as it builds on existing IML methods.
The paper tackled the problem of enabling non-experts to teach agents via verbal instructions in interactive machine learning, and found that their Newtonian Action Advice method outperformed the state-of-the-art Policy Shaping in both cumulative reward and human frustration metrics.
A goal of Interactive Machine Learning (IML) is to enable people without specialized training to teach agents how to perform tasks. Many of the existing machine learning algorithms that learn from human instructions are evaluated using simulated feedback and focus on how quickly the agent learns. While this is valuable information, it ignores important aspects of the human-agent interaction such as frustration. In this paper, we present the Newtonian Action Advice agent, a new method of incorporating human verbal action advice with Reinforcement Learning (RL) in a way that improves the human-agent interaction. In addition to simulations, we validated the Newtonian Action Advice algorithm by conducting a human-subject experiment. The results show that Newtonian Action Advice can perform better than Policy Shaping, a state-of-the-art IML algorithm, both in terms of RL metrics like cumulative reward and human factors metrics like frustration.