### Abstract

This paper develops a general policy for learning relevant features of an imitation task. We restrict our study to imitation of manipulative tasks or of gestures. The imitation process is modeled as a hierarchical optimization system, which minimizes the discrepancy between two multi-dimensional datasets. To classify across manipulation strategies, we apply a probabilistic analysis to data in Cartesian and joint spaces. We determine a general metric that optimizes the policy of task reproduction, following strategy determination. The model successfully discovers strategies in six different imitative tasks and controls task reproduction by a full body humanoid robot.

### Bibtex reference

@article{Billard_et_al04,
author = "Billard, A. and Epars, Y. and Calinon, S. and Cheng, G. and Schaal, S.",
title = "Discovering Optimal Imitation Strategies",
journal = "Robotics and Autonomous Systems",
year = "2004",
volume="47",
number="2-3",
pages="69--77"
}