A Metric for the Evaluation of Imitation

Amit Ramesh

Learning by imitation and learning from demonstration have recently received considerable attention in robotics. However, very little research has been in the direction of providing a quantitative measure of the quality of imitation. We are interested in developing a generalized comprehensive metric that provides a scalar measure of dissimilarity/distance between any pair of action sequences. By developing such a metric, we intend to provide a standardized means to quantitatively evaluate imitation.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.