Titre : Value-sensitive knowledge evolution

Sujet proposé dans : M2 MOSIG, Projet --- M2 MSIAM, Projet --- M2R Informatique, Projet --- Magistere, M2

Responsable(s) :

Mots-clés : Knowledge evolution, Artificial intelligence, Multi-agent system, Knowledge representation, Value-sensitive design
Durée du projet : 5 mois, possibilité de continuer en thèse
Nombre maximal d'étudiants : 1
Places disponibles : 1
Interrogation effectuée le : 24 avril 2024, à 12 heures 04


Description

A

Agents may evolve their knowledge using various kinds of knowledge adaptation operators. However, the choice of an operator is not well-justified. We plan to guide the choice of operators with general values held by agents.

Cultural evolution is the application of evolution theory to culture [Messoudi 2006]. It may be addressed through multi-agent simulations [Steels 2012]. Experimental cultural evolution provides a population of agents with interaction games that are played randomly. In reaction to the outcome of such games, agents adapt their knowledge. It is possible to test hypotheses by precisely crafting the rules used by agents in games and observing the consequences.

Our ambition is to understand and develop general mechanisms by which a society evolves its knowledge. For that purpose, we adapted this approach to the evolution of the way agents represent knowledge [Anslow & Rovatsos, 2015; Chocron & Schorlemmer, 2016; Euzenat, 2017; Bourahla et al., 2021]. Agents would communicate to perform a task that involve their knowledge and adapt it through adaptation operators when communication fails. We showed that cultural repair is able to converge towards successful communication and improves the objective correctness of knowledge.

However, people base their attitudes and behaviour on deeper ground than the choice of technical operators [Vanhée, 2015]. Values are such a basis. Values are generic indicators (e.g. timeliness, obedience, achievement, power) on which individuals rely for distinguising ``good" from ``bad", desirable from non-desirable. Individuals may use their values in order to guide their long-term attitude and short-term behaviour.

Hence, they may in the short term evaluate their situation and take action according to these values. On the longer term, they may elect specific operators based on their values. Ultimately, they may design value-driven operators.

This master topic aims at considering two questions: (a) how may the choice of particular operators and modalities be guided by values held by agents, and (b) what consequences have such choices of values. In particular, it may also investigate the connection between the short-term and longer term use of values in this context.

For that purpose, the candidate can concentrate on the core model of values introduced by Schwartz [Schwartz, 2012] and study the way values can guide the behaviour of agents through its choice of repair operators. Then, experiments in which groups of agents play the same games according to different values may be performed to observe the outcome in terms of agreement speed or quality of the resulting knowledge.

This work is part of an ambitious program towards what we call cultural knowledge evolution. It is part of the MIAI Knowledge communication and evolution chair and as such may lead to a PhD thesis.

References:

[Anslow & Rovatsos, 2015] Michael Anslow, Michael Rovatsos, Aligning experientially grounded ontologies using language games, Proc. 4th international workshop on graph structure for knowledge representation, Buenos Aires (AR), pp15-31, 2015 [DOI:10.1007/978-3-319-28702-7_2]
[Bourahla et al., 2021] Yasser Bourahla, Manuel Atencia, Jérôme Euzenat, Knowledge improvement and diversity under interaction-driven adaptation of learned ontologies, Proc. 20th ACM international conference on Autonomous Agents and Multi-Agent Systems (AAMAS), London (UK), pp242-250, 2021 http://www.ifaamas.org/Proceedings/aamas2021/pdfs/p242.pdf
[Chocron & Schorlemmer, 2016] Paula Chocron, Marco Schorlemmer, Attuning ontology alignments to semantically heterogeneous multi-agent interactions, Proc. 22nd European Conference on Artificial Intelligence, Der Haague (NL), pp871-879, 2016 [DOI:10.3233/978-1-61499-672-9-871]
[Euzenat & Shvaiko, 2013] Jérôme Euzenat, Pavel Shvaiko, Ontology matching, 2nd edition, Springer-Verlag, Heildelberg (DE), 2013
[Euzenat, 2017] Jérôme Euzenat, Communication-driven ontology alignment repair and expansion, in: Proc. 26th International joint conference on artificial intelligence (IJCAI), Melbourne (AU), pp185-191, 2017 ftp://ftp.inrialpes.fr/pub/moex/papers/euzenat2017a.pdf
[Mesoudi et al., 2006] Alex Mesoudi, Andrew Whiten, Kevin Laland, Towards a unfied science of cultural evolution, Behavioral and brain sciences 29(4):329–383, 2006 http://alexmesoudi.com/s/Mesoudi_Whiten_Laland_BBS_2006.pdf
[Schwartz, 2012] Shalom Schwartz, An Overview of the Schwartz Theory of Basic Values, Online readings in psychology and culture 2(1) https://doi.org/10.9707/2307-0919.1116
[Steels, 2012] Luc Steels (ed.), Experiments in cultural language evolution, John Benjamins, Amsterdam (NL), 2012
[Vanhée, 2015] Loïs Vanhée, Using culture and values to support flexible coordination, Phd dissertation, Utrecht university (NL), 2015

Links: