Online goal recognition through mirroring: humans and agents

Mor Vered, Gal A. Kaminka, Sivan Biham

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review


Goal recognition is the problem of inferring the (unobserved) goal of an agent, based on a sequence of its observed actions. Inspired by mirroring processes in human brains, we advocate goal mirroring, an online recognition approach that uses a black-box planner to generate recognition hypotheses. This approach avoids the prevalent assumption in current approaches, which rely on a dedicated plan library, representing all known plans to achieve known goals. Such methods are inherently limited to the knowledge represented in the library. In this paper, we (i) describe a novel online goal mirroring algorithm for continuous spaces; (ii) evaluate a novel heuristic for choosing between competing recognition hypotheses; (iii) contrast machine and human recognition in two challenging domains, revealing insights as to human capabilities; and (iv) compare mirroring to library-based methods.
Original languageEnglish
Title of host publicationFourth Annual Conference on Advances in Cognitive Systems
Subtitle of host publicationJune 23-26, 2016, Evanston, Illinois
EditorsKenneth Forbus, Tom Hinrichs, Carrie Ost
Place of PublicationKirkland WA USA
PublisherCognitive Systems Foundation
Number of pages13
Publication statusPublished - 2016
Externally publishedYes
EventAnnual Conference on Advances in Cognitive Systems 2016 - Evanston, United States of America
Duration: 23 Jun 201626 Jun 2016
Conference number: 4th

Publication series

NameAdvances in Cognitive Systems
ISSN (Print)2324-8416


ConferenceAnnual Conference on Advances in Cognitive Systems 2016
Country/TerritoryUnited States of America
Internet address

Cite this