Learning Within the BDI Framework: An Empirical Analysis
One of the limitations of the BDI (Belief-Desire-Intention) model is the lack of any explicit mechanisms within the architecture to be able to learn. In particular, BDI agents do not possess the ability to adapt based on past experience. This is important in dynamic environments since they can change, causing methods for achieving goals that worked well previously to become inefficient or ineffective. We present a model in which learning can be utilised by a BDI agent and verify this model experimentally using two learning algorithms.
KeywordsSearch Space Analogous Reasoning Inductive Logic Programming Knowledge Extractor High Order Function
Unable to display preview. Download preview PDF.
- 2.Wooldridge, M.: An Introduction To MultiAgent Systems, 1st edn. John Wiley and Sons Ltd, Chichester (2002)Google Scholar
- 3.Bratman, M.E.: Intentions, Plans, and Practical Reason, 1st edn. Harvard University Press, Cambridge (1987)Google Scholar
- 4.Rao, A.S., Georgeff, M.P.: BDI-Agents: From Theory to Practice. In: Proceedings of the First International Conference on Multiagent Systems (1995)Google Scholar
- 6.Ng, K.S.: Alkemy: A Learning System Based on an Expressive Knowledge Representation (2004), Available from http://users.rsise.anu.edu.au/~kee/Alkemy/
- 8.Olivia, C., Chang, C.F., Enguix, C.F., Ghose, A.K.: Case-Based BDI Agents: An Effective Approach for Intelligent Search on the World Wide Web. In: AAAI Spring Symposium (1999)Google Scholar
- 10.Veloso, M., Carbonell, J., Perez, A., Borrajo, D., Fink, E., Blythe, J.: Integrating Planning and Learning: The PRODIGY Architecture. Journal of Experimental and Theoretical Artificial Intelligence 7(1) (1995)Google Scholar