Incorporating observer biases in keyhole plan recognition (efficiently!)

Dorit Avrahami-Zilberbrand, Gal A. Kaminka

Research output: Contribution to journalArticlepeer-review

28 Scopus citations

Abstract

Plan recognition is the process of inferring other agents' plans and goals based on their observable actions. Essentially all previous work in plan recognition has focused on the recognition process itself, with no regard to the use of the information in the recognizing agent. As a result, low-likelihood recognition hypotheses that may imply significant meaning to the observer, are ignored in existing work. In this paper, we present novel efficient algorithms that allows the observer to incorporate her own biases and preferences - in the form of a utility function - into the plan recognition process. This allows choosing recognition hypotheses based on their expected utility to the observer. We call this Utility-based Plan Recognition (UPR). While reasoning about such expected utilities is intractable in the general case, we present a hybrid symbolic/decision-theoretic plan recognizer, whose complexity is O(NDT), where N is the plan library size, D is the depth of the library and T is the number of observations. We demonstrate the efficacy of this approach with experimental results in several challenging recognition tasks. Copyright © 2007, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
Original languageEnglish
JournalProceedings of the National Conference on Artificial Intelligence
Volume2
StatePublished - 28 Nov 2007

Fingerprint

Dive into the research topics of 'Incorporating observer biases in keyhole plan recognition (efficiently!)'. Together they form a unique fingerprint.

Cite this