Timezone: »

Showing versus doing: Teaching by demonstration
Mark Ho · Michael Littman · James MacGlashan · Fiery Cushman · Joe Austerweil · Joseph L Austerweil

Wed Dec 07 09:00 AM -- 12:30 PM (PST) @ Area 5+6+7+8 #198

People often learn from others' demonstrations, and classic inverse reinforcement learning (IRL) algorithms have brought us closer to realizing this capacity in machines. In contrast, teaching by demonstration has been less well studied computationally. Here, we develop a novel Bayesian model for teaching by demonstration. Stark differences arise when demonstrators are intentionally teaching a task versus simply performing a task. In two experiments, we show that human participants systematically modify their teaching behavior consistent with the predictions of our model. Further, we show that even standard IRL algorithms benefit when learning from behaviors that are intentionally pedagogical. We conclude by discussing IRL algorithms that can take advantage of intentional pedagogy.

Author Information

Mark Ho (Brown University)
Michael Littman (Brown University)
James MacGlashan (Brown University)
Fiery Cushman (Harvard University)

Fiery Cushman is Assistant Professor of Psychology at Harvard University, where he directs the Moral Psychology Research Laboratory.  His research investigates the cognitive mechanisms responsible for human moral judgment, along with their development, evolutionary history and neural basis.  His work often draws from classic philosophical dilemmas, and has focused in particular on the psychology of punishment and the aversion to harmful action.  He received his BA and PhD from Harvard University, where he also completed a post-doctoral fellowship. He served as Assistant Professor of Cognitive, Linguistic and Psychological Sciences at Brown University from 2011 to 2014.

Joe Austerweil (University of Wisconsin-Madison)
Joseph L Austerweil (University of Wisconsin-Madison)

As a computational cognitive psychologist, my research program explores questions at the intersection of perception and higher-level cognition. I use recent advances in statistics and computer science to formulate ideal learner models to see how they solve these problems and then test the model predictions using traditional behavioral experimentation. Ideal learner models help us understand the knowledge people use to solve problems because such knowledge must be made explicit for the ideal learner model to successfully produce human behavior. This method yields novel machine learning methods and leads to the discovery of new psychological principles.

More from the Same Authors