Quantifying Teaching Behaviour in Robot Learning from Demonstration

TL;DR

This paper introduces a model to evaluate and enhance teaching effectiveness in robot learning from demonstration, demonstrating significant efficiency improvements through guided feedback.

Contribution

It presents a novel framework that models the teacher's understanding and influence, enabling better evaluation and improvement of teaching in robot learning from demonstration.

Findings

Teachers struggle with providing effective demonstrations.

Evaluation and feedback can improve teaching efficiency by approximately 169-180%.

The proposed model clarifies teaching objectives and failure modes.

Abstract

Learning from demonstration allows for rapid deployment of robot manipulators to a great many tasks, by relying on a person showing the robot what to do rather than programming it. While this approach provides many opportunities, measuring, evaluating and improving the person's teaching ability has remained largely unexplored in robot manipulation research. To this end, a model for learning from demonstration is presented here which incorporates the teacher's understanding of, and influence on, the learner. The proposed model is used to clarify the teacher's objectives during learning from demonstration, providing new views on how teaching failures and efficiency can be defined. The benefit of this approach is shown in two experiments (N=30 and N=36, respectively), which highlight the difficulty teachers have in providing effective demonstrations, and show how ~169-180% improvement in teaching efficiency can be achieved through evaluation and feedback shaped by the proposed framework, relative to unguided teaching.