How Humans Teach Agents: A New Experimental Perspective (2012)
W. Bradley Knox and Brian D. Glass and Bradley C. Love and W. Todd Maddox and Peter Stone
Human beings are a largely untapped source of in-the-loop knowledge and guidance for computational learning agents, including robots. To effectively design agents that leverage available human expertise, we need to understand how people naturally teach. In this paper, we describe two experiments that ask how differing conditions affect a human teacher's feedback frequency and the computational agent's learned performance. The first experiment considers the impact of a self-perceived teaching role in contrast to believing one is merely critiquing a recording. The second considers whether a human trainer will give more frequent feedback if the agent acts less greedily (i.e., choosing actions believed to be worse) when the trainer's recent feedback frequency decreases. From the results of these experiments, we draw three main conclusions that inform the design of agents. More broadly, these two studies stand as early examples of a nascent technique of using agents as highly specifiable social entities in experiments on human behavior.
International Journal of Social Robotics, 4:409-421, 2012. Springer Netherlands.

W. Bradley Knox bradknox [at] mit edu
Peter Stone pstone [at] cs utexas edu