Detecting ‘When to Imitate’ in a Social Context with a Human Caregiver

Abstract

Research on imitation learning is guided primarily by five questions that need to be decided on by a robot: who to imitate, when to imitate, what to imitate, how to imitate, and how to evaluate a successful imitation [4]. There has been intensive research on solving these questions by technical approaches (cf., e.g., [1]), concentrating mainly