Gal A. Kaminka: Publications

Sorted by DateClassified by Publication TypeClassified by TopicGrouped by Student (current)Grouped by Former Students

Plan Recognition in Continuous Domains

Gal A. Kaminka, Mor Vered, and Noa Agmon. Plan Recognition in Continuous Domains. In Proceedings of the AAAI Conference on Artificial Intelligence , 2018.
An earlier version was published in the IJCAI 2017 workshop on Goal Reasoning.

Download

[PDF]598.6kB  

Abstract

Plan recognition is the task of inferring the plan of an agent, basedon an incomplete sequence of its observed actions. Previous formulations of planrecognition commit early to discretizations of the environment and the observedagent's actions. This leads to reduced recognition accuracy. To address this, wefirst provide a formalization of recognition problems which admits continuousenvironments, as well as discrete domains. We then show that throughmirroring---generalizing plan-recognition by planning---wecan apply continuous-world motion planners in plan recognition.We provide formal arguments for the usefulness of mirroring, and empiricallyevaluate mirroring in more than a thousand recognition problemsin three continuous domains and six classical planning domains.

Additional Information

BibTeX

@inproceedings{aaai18,
	author = {Gal A. Kaminka and Mor Vered and Noa Agmon},
title = {Plan Recognition in Continuous Domains},
booktitle = AAAI,
OPTcrossref = {crossref},
OPTkey = {key},
OPTpages = {pages},
year = {2018},
OPTeditor = {editor},
OPTvolume = {volume},
OPTnumber = {number},
OPTseries = {series},
OPTaddress = {address},
OPTmonth = {month},
OPTorganization = {organization},
OPTpublisher = {publisher},
wwwnote = {An earlier version was published in the IJCAI 2017 workshop on Goal Reasoning.},
OPTannote = {annote},
OPTnote = {},
OPTkeywords = {},
abstract = {
Plan  recognition is the task of inferring the  plan of an agent, based
on an incomplete sequence of its observed actions. Previous formulations of plan
recognition commit early to discretizations of the environment and the observed
agent's actions. This leads to reduced recognition accuracy. To address this, we
first provide a formalization of recognition problems which admits continuous
environments, as well as discrete domains. We then show that through
\textit{mirroring}---generalizing plan-recognition by planning---we
can apply continuous-world motion planners in plan recognition.
We provide formal arguments for the usefulness of mirroring, and empirically
evaluate mirroring  in more than a thousand recognition problems
in three continuous domains and six classical planning domains.
},
}

Generated by bib2html.pl (written by Patrick Riley ) on Fri Aug 30, 2024 17:29:51