Definition
Model-based imitation refers to a family of machine-learning methods, which can be used to quickly generate a rough solution to a given control task, usually in robotics, using demonstrated behavior. The premise is that a large class of tasks can be demonstrated, either by a human, e.g., household tasks for domestic robots, or by other “teacher” robots that are more skilled than the learner. The solution, as observed by the learner, typically consists of trajectories in the space of some observed variables. The appropriate actions that are needed to steer the robot along the trajectories usually cannot be observed, but they can be learnt by using a model of the robot’s dynamics. This gives rise to the term model-based imitation.
Theoretical Background
It has long been known that humans and animals use imitation as a mechanism for acquiring knowledge. Consequently, algorithms...
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Atkeson, C., & Schaal, S. (1997). Robot learning from demonstration (pp. 12–20). San Francisco: Morgan Kaufmann.
Bratko, I., & Šuc, D. (2002). Using machine learning to understand operator’s skill. In T. Hendtlass & M. Ali (Eds.), Developments in applied artificial intelligence, (Lecture notes in computer science, Lecture notes in artificial intelligence, 2358) (pp. 812–823). Berlin: Springer.
Kuniyoshi, Y., Inaba, M., & Inoue, H. (1994). Learning by watching: Extracting reusable task knowledge from visual observation of human performance. IEEE Transactions on Robotics and Automation, 10(6), 799–822.
Schaal, S. (1997). Robot learning from demonstration. In D. H. Fisher Jr. (Ed.), International conference on machine learning (ICML97) (pp. 12–20). San Francisco: Morgan Kaufmann. 1997.
Schaal, S. (1999). Is imitation learning the route to humanoid robots? Trends in Cognitive Sciences, 3(6), 233–242. June 1999.
Peters, J., Vijayakumar, S., & Schaal, S. (2003). Reinforcement Learning for Humanoid Robotics. Proceedings of the Third IEEE-RAS International Conference on Humanoid Robots (Humanoids2003), 1–20 (September 2003).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer Science+Business Media, LLC
About this entry
Cite this entry
Babuska, R. (2012). Model-Based Imitation Learning. In: Seel, N.M. (eds) Encyclopedia of the Sciences of Learning. Springer, Boston, MA. https://doi.org/10.1007/978-1-4419-1428-6_563
Download citation
DOI: https://doi.org/10.1007/978-1-4419-1428-6_563
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4419-1427-9
Online ISBN: 978-1-4419-1428-6
eBook Packages: Humanities, Social Sciences and Law