Learning action effects in partially observable domains

Kira Mourao, Ron Petrick, Mark Steedman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

We investigate the problem of learning action effects in partially observable STRIPS planning domains. Our approach is based on a voted kernel perceptron learning model, where action and state information is encoded in a compact vector representation as input to the learning mechanism, and resulting state changes are produced as output. Our approach relies on deictic features that assume an attentional mechanism that reduces the size of the representation. We evaluate our approach on a number of partially observable planning domains, and show that it can quickly learn the dynamics of such domains, with low average error rates. We show that our approach handles noisy domains, conditional effects, and that it scales independently of the number of objects in a domain.
Original languageEnglish
Title of host publicationECAI 2010 - 19th European Conference on Artificial Intelligence, Lisbon, Portugal, August 16-20, 2010, Proceedings.
EditorsHelder Coelho, Rudi Studer, Michael Wooldridge
PublisherIOS Press
Pages973-974
Number of pages2
ISBN (Print)978-1-60750-605-8
DOIs
Publication statusPublished - 2010

Fingerprint

Dive into the research topics of 'Learning action effects in partially observable domains'. Together they form a unique fingerprint.

Cite this