Abstract
Integrating different aspects of learning in planning systems has been one of the most promising trends towards solving the problem of complexity and scalability that affects the planning systems. Most of the times, the user has been put aside this integration, as most approaches consider the existence of per-fect models of the domain to act as an oracle for the ability of the system to learn, either by observation or practice. We have decided to put the user in the learning loop, in a mixed initiative approach to online learning. The user will propose problems and guide the system through the space of possible alterna-tives to solve it, therefore guiding learning. To generate learning episodes, the system also performs its own experimentation, relying on the user to classify the resulting scenarios, allowing some otherwise impossible learning approaches. The user is not a perfect model, so mistakes can occur, during this interaction. We have to be prepared to cope with these imperfections, and recover from user induced errors.
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
Preview
Unable to display preview. Download preview PDF.
References
Borrajo, D., Veloso, M.: Incremental Learning of Control Knowledge for Nonlinear Problem Solving. In: Bergadano, F., De Raedt, L. (eds.) ECML 1994. LNCS, vol. 784, pp. 64–82. Springer, Heidelberg (1994)
Carbonell, J., Gil, Y.: Learning by Experimentation: The Operator Refinement Method. In: Michalsky, R.S., Kodratoff, Y. (eds.) Machine Learning: An Artificial Intelligent Approach, vol. III, pp. 191–213. Morgan Kaufmann, San Francisco (1990)
Cheng, P., Carbonell, J.: The FERMI System: Inducing Iterative Macro-operators from Experience. In: AAAI 1986, pp. 490–495 (1986)
Ferreira, J.L., Costa, E.: Learning Operators While Planning. In: Proceedings of the Sixth Portuguese Conference on Artificial Intelligence, pp. 311–323 (1993)
Ferreira, J.L., Costa, E.: Opportunities for Learning in Planning. In: Proceedings of the XI Brazilian Symposium on Artificial Intelligence, pp. 321–334 (1994)
Huffman, S., Laird, J.: Flexible Instructable Agents. Journal of Applied AI Research 3, 271–324 (1996)
Kelleher, G., Cohn, A.G.: Automatically Synthesising Domain Constraints from Operator Descriptions. In: ECAI 1992, pp. 653–655 (1992)
Kodratoff, Y., Tecuci, G.: DISCIPLE-I: Interactive Apprentice System in Weak Theory Fields. In: IJCAI 1991 (1991)
Laird, J.E., Newell, A., Rosenbloom, P.S.: Soar: An architecture for general intelligence. Artificial Intelligence 33(1), 1–64 (1987)
Langley, P., Allen, J.: The Acquisition of Human Planning Expertise. In: Proceedings of the 8th International Workshop on Machine Learning, pp. 80–84 (1991)
Leckie, C., Zuckerman, I.: An Inductive Approach to Learning Search Control Rules for Planning. In: IJCAI 1993, pp. 1100–1105 (1993)
Macedo, L., Pereira, F., Grilo, C., Cardoso, A.: Plan Cases as Structured Networks of Hierarchical and Temporal Related Case Pieces. In: EWCBR 1996. LNCS, vol. 1168. Springer, Heidelberg (1996)
Mitchell, T.M.: Version Spaces: An Approach to Concept Learning. PhD Thesis, Stanford University (1978)
Veloso, M., Carbonell, J., Pérez, A., Borrajo, D., Fink, E., Blythe, J.: Integrating Planning and Learning: The PRODIGY Architecture. Journal of Experimental and Theoretical Artificial Intelligence 7(1), 81–120 (1995)
Wang, X.: Learning Planning Operators by Observation and Practice, PhD Thesis, School of Computer Science, Carnegie Mellon University, CMU-CS-96-154 (1996)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ferreira, J.L., Costa, E.J. (1998). Planning and Learning: Put the User in the Loop. In: de Oliveira, F.M. (eds) Advances in Artificial Intelligence. SBIA 1998. Lecture Notes in Computer Science(), vol 1515. Springer, Berlin, Heidelberg. https://doi.org/10.1007/10692710_11
Download citation
DOI: https://doi.org/10.1007/10692710_11
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65190-1
Online ISBN: 978-3-540-49523-9
eBook Packages: Springer Book Archive