Cultural knowledge evolution considers how agents evolve their knowledge through communicating with each others. However, these agents evolve within distinct populations. The questions of how such populations split and merge and how this influences agent knowledge are raised.
Cultural evolution is the application of evolution theory to culture. It has been applied to various aspects of our life in societies: from customs to languages, from boat shapes to company structures [Messoudi, 2011].
Cultural knowledge evolution deals with the evolution of knowledge representation in a group of computer agents. For that purpose, cooperating agents adapt their knowledge to the situations they are exposed to and the feedback they receive from others. After playing this game repeatedly, it is possible to observe the properties of the resulting knowledge. This framework has been considered in the context of evolving natural languages [Steels, 2012]. We have applied it to ontology alignment repair, i.e. the improvement of incorrect alignments [Euzenat, 2017] and ontology evolution [Bourahla et al., 2021]. We have shown that it converges towards successful communication, improves the intrinsic knowledge quality but preserves the diversity of agent knowledge.
This PhD proposal focusses on the behaviour of populations of agents as opposed to agents individually. The population approach is central to evolution theory. Moreover, what defines culture is not what is held by a single agent, but what is shared among a group of agents. Hence we want to investigate the connection between populations and knowledge. This requires to answer several questions which are as many population design options.
The first question is what does make a population. It can be geography, interbreeding, the topology of agent connections, the shared culture (here knowledge) or the capability to communicate.
Once understood what can be considered as a population, it is necessary to consider what is shared within such a population. On one end of a spectrum, this may be nothing, a population is thus simply the privileged, but not exclusive, interaction space. Hence, the population would simply provide stronger social pressure. On the other end, it may be that agents of the same population share the same culture, here they share their knowledge, which is then difficult to evolve. Between these two ends, agents may have specific means to synchronise their knowledge, they may have the same language, which facilitates their communications, or they may share common values, which they use for evolving their knowledge. They may also share common motivations. Such different population-related modalities constrain agent behaviours.
How the knowledge of an agents impacts that of the population it belongs to? Agents may hold their knowledge from the populations they belong to, or may determine the knowledge of the populations they belong to. If knowledge has an influence on the definition of populations, then interacting should have an impact on population composition which in turn will certainly have an impact on agent knowledge. There is then co-evolution of knowledge and populations, as was already observed in [Axelrod, 1997].
This would also impact how populations are born and die, or, more concretely, how they split, due to diverging knowledge, or merge, due to converging knowledge. From a knowledge standpoint, this relates to how a population, interacting agent by agent, adopts the knowledge of another. It can also be how sub populations emerge due to divergence in the knowledge the agents hold.
Finally, the goal of this topic is to understand the possible influence of these different settings to knowledge, or more precisely to the properties of a population's knowledge. Do some of these conditions lead to more homogeneous knowledge, i.e. less diversity? Does this homogeneity has impact on future evolution of the knowledge held within a population? In turn, does it affect the correctness of resulting knowledge? Is there a trade-off between correctness and diversity?
In summary, how the answers to these questions impact the propagation and evolution of knowledge among agents. These problems may be treated both theoretically and experimentally.
References:
[Axelrod, 1997] Robert Axelrod, The dissemination of culture: a model with local convergence and global polarization, Journal of conflict resolution 41:203–226, 1997.
[Bourahla, 2021] Yasser Bourahla, Manuel Atencia, Jérôme Euzenat, Knowledge improvement and diversity under interaction-driven adaptation of learned ontologies, Proc. 20th AAMAS, London (UK), pp242-250, 2021 https://moex.inria.fr/files/papers/bourahla2021a.pdf
[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 https://moex.inria.fr/files/papers/euzenat2017a.pdf
[Mesoudi, 2011] Alex Mesoudi, Cultural evolution: how Darwinian theory can explain human culture and synthesize the social sciences, Chicago university press, Chicago (IL US), 2011 See also: 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
[Steels, 2012] Luc Steels (ed.), Experiments in cultural language evolution, John Benjamins, Amsterdam (NL), 2012
Links:
Qualification: Master or equivalent in computer science.
Researched skills:
Doctoral school: MSTII, Université Grenoble Alpes.
Advisor: Jérôme Euzenat (Jerome:Euzenat#inria:fr).
Group: The work will be carried out in the mOeX team common to INRIA & LIG. mOeX is dedicated to study knowledge evolution through adaptation. It gathers researchers which have taken an active part these past 15 years in the development of the semantic web and more specifically ontology matching and data interlinking.
Place of work: The position is located at INRIA Montbonnot (near Grenoble) a main computer science research lab, in a stimulating research environment.
Hiring date: October 2024.
Duration: 36 months
Salary: 1982€/month (gross, before social contributions and taxes).
Deadline: as soon as possible.
Contact: For further information, contact us.
Procedure: Contact us .
File: Provide Vitæ, motivation letter and references. It is very good if you can provide a Master report and we will ask for your marks in Master, so if you have them, you can join them.