Online probabilistic goal recognition over nominal models

Ramon Fraga Pereira, Mor Vered, Felipe Meneguzzi, Miquel Ramirez

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

Abstract

This paper revisits probabilistic, model--based goal recognition to study the implications of the use of nominal models to estimate the posterior probability distribution over a finite set of hypothetical goals. Existing model--based approaches rely on expert knowledge to produce symbolic descriptions of the dynamic constraints domain objects are subject to, and these are assumed to produce correct predictions. We abandon this assumption to consider the use of nominal models that are learnt from observations on transitions of systems with unknown dynamics. Leveraging existing work on the acquisition of domain models via deep learning for hybrid planning we adapt and evaluate existing goal recognition approaches to analyse how prediction errors, inherent to system dynamics identification and model learning techniques have an impact over recognition error rates.
Original languageEnglish
Title of host publicationProceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence
EditorsSarit Kraus
Place of PublicationCalifornia USA
PublisherInternational Joint Conferences on Artificial Intelligence
Pages5547-5553
Number of pages7
ISBN (Electronic)9780999241141
DOIs
Publication statusPublished - 2019
EventInternational Joint Conference on Artificial Intelligence 2019 - Macao, China
Duration: 10 Aug 201916 Aug 2019
Conference number: 28th
https://ijcai19.org/

Conference

ConferenceInternational Joint Conference on Artificial Intelligence 2019
Abbreviated titleIJCAI-19
CountryChina
CityMacao
Period10/08/1916/08/19
Internet address

Keywords

  • Goal Recognition
  • Online

Cite this

Pereira, R. F., Vered, M., Meneguzzi, F., & Ramirez, M. (2019). Online probabilistic goal recognition over nominal models. In S. Kraus (Ed.), Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence (pp. 5547-5553). California USA: International Joint Conferences on Artificial Intelligence. https://doi.org/10.24963/ijcai.2019/770
Pereira, Ramon Fraga ; Vered, Mor ; Meneguzzi, Felipe ; Ramirez, Miquel. / Online probabilistic goal recognition over nominal models. Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. editor / Sarit Kraus. California USA : International Joint Conferences on Artificial Intelligence, 2019. pp. 5547-5553
@inproceedings{d16610d5fe4a4dcc84c133e0da2f61f5,
title = "Online probabilistic goal recognition over nominal models",
abstract = "This paper revisits probabilistic, model--based goal recognition to study the implications of the use of nominal models to estimate the posterior probability distribution over a finite set of hypothetical goals. Existing model--based approaches rely on expert knowledge to produce symbolic descriptions of the dynamic constraints domain objects are subject to, and these are assumed to produce correct predictions. We abandon this assumption to consider the use of nominal models that are learnt from observations on transitions of systems with unknown dynamics. Leveraging existing work on the acquisition of domain models via deep learning for hybrid planning we adapt and evaluate existing goal recognition approaches to analyse how prediction errors, inherent to system dynamics identification and model learning techniques have an impact over recognition error rates.",
keywords = "Goal Recognition, Online",
author = "Pereira, {Ramon Fraga} and Mor Vered and Felipe Meneguzzi and Miquel Ramirez",
year = "2019",
doi = "10.24963/ijcai.2019/770",
language = "English",
pages = "5547--5553",
editor = "Sarit Kraus",
booktitle = "Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence",
publisher = "International Joint Conferences on Artificial Intelligence",

}

Pereira, RF, Vered, M, Meneguzzi, F & Ramirez, M 2019, Online probabilistic goal recognition over nominal models. in S Kraus (ed.), Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. International Joint Conferences on Artificial Intelligence, California USA, pp. 5547-5553, International Joint Conference on Artificial Intelligence 2019, Macao, China, 10/08/19. https://doi.org/10.24963/ijcai.2019/770

Online probabilistic goal recognition over nominal models. / Pereira, Ramon Fraga; Vered, Mor; Meneguzzi, Felipe; Ramirez, Miquel.

Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. ed. / Sarit Kraus. California USA : International Joint Conferences on Artificial Intelligence, 2019. p. 5547-5553.

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

TY - GEN

T1 - Online probabilistic goal recognition over nominal models

AU - Pereira, Ramon Fraga

AU - Vered, Mor

AU - Meneguzzi, Felipe

AU - Ramirez, Miquel

PY - 2019

Y1 - 2019

N2 - This paper revisits probabilistic, model--based goal recognition to study the implications of the use of nominal models to estimate the posterior probability distribution over a finite set of hypothetical goals. Existing model--based approaches rely on expert knowledge to produce symbolic descriptions of the dynamic constraints domain objects are subject to, and these are assumed to produce correct predictions. We abandon this assumption to consider the use of nominal models that are learnt from observations on transitions of systems with unknown dynamics. Leveraging existing work on the acquisition of domain models via deep learning for hybrid planning we adapt and evaluate existing goal recognition approaches to analyse how prediction errors, inherent to system dynamics identification and model learning techniques have an impact over recognition error rates.

AB - This paper revisits probabilistic, model--based goal recognition to study the implications of the use of nominal models to estimate the posterior probability distribution over a finite set of hypothetical goals. Existing model--based approaches rely on expert knowledge to produce symbolic descriptions of the dynamic constraints domain objects are subject to, and these are assumed to produce correct predictions. We abandon this assumption to consider the use of nominal models that are learnt from observations on transitions of systems with unknown dynamics. Leveraging existing work on the acquisition of domain models via deep learning for hybrid planning we adapt and evaluate existing goal recognition approaches to analyse how prediction errors, inherent to system dynamics identification and model learning techniques have an impact over recognition error rates.

KW - Goal Recognition

KW - Online

U2 - 10.24963/ijcai.2019/770

DO - 10.24963/ijcai.2019/770

M3 - Conference Paper

SP - 5547

EP - 5553

BT - Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence

A2 - Kraus, Sarit

PB - International Joint Conferences on Artificial Intelligence

CY - California USA

ER -

Pereira RF, Vered M, Meneguzzi F, Ramirez M. Online probabilistic goal recognition over nominal models. In Kraus S, editor, Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. California USA: International Joint Conferences on Artificial Intelligence. 2019. p. 5547-5553 https://doi.org/10.24963/ijcai.2019/770