Postdoc position in Persuasion in Hybrid Intelligence Systems

A 3-year postdoc position in artificial intelligence is available at the Bernoulli Institute of the University of Groningen in the Netherlands. The position is situated in the multi-agent systems group as part of the Hybrid Intelligence Centre, under the joint supervision of Dr. Harmen de Weerd, Prof.dr. Rineke Verbrugge, and Dr. Max van Duijn (Leiden University).

Applicants must have a PhD in a computational field such as Artificial Intelligence or Computational Cognitive Science, or a strong background in experimental behavioral research. Candidates should have strong written and spoken English and be able to collaborate effectively with people from different backgrounds. Moreover, they should have demonstrated ability to publish and present at international venues. In addition, a successful applicant has good programming skills, a keen interest in human-machine communication, and is motivated towards contributing to Hybrid Intelligence and bridging the gap between human and machine.

The successful candidate has a strong background in experimental research, good programming skills, a keen interest in human-machine communication, and is motivated towards contributing to Hybrid Intelligence.

You are also expected to interact scientifically with other research staff and students, as well as organize and participate in seminars, such as those of the Hybrid Intelligence Centre. The research is also expected to be performed within the context of the Hybrid Intelligence Center.

The position comes with mild teaching duties (in English, about 10% of the appointment). Knowledge of the Dutch language is not required.

The research focus is on persuasion in Hybrid Intelligence. In true hybrid intelligent teams, artificial agents may have a viewpoint or goal that is different from that of a (human) collaborator. In such situations, the artificial agent should aim to align these viewpoints or goals by adjusting those of their collaborator. This project revolves around the question of how persuasion of human agents by computational agents can work in a one-to-one setting, and in a way that is satisfactory in terms of outcome as well as the subjective experience of the human collaborator. Importantly, the type of persuasion of interest is overt, rather than covert methods such as nudging.

As part of the research, concrete evaluation is envisioned through the virtual museum guide developed by the national Hybrid Intelligence Centre (https://www.hybrid-intelligence-centre.nl/), where an artificial museum guide that is knowledgeable about the exhibits presents visitors with information that is personally relevant to the visitor, but may also present new perspectives on art.

Within the Faculty of Science and Engineering (FSE), the Bernoulli Institute conducts research combining theoretical as well as practical aspects of three disciplines, namely Mathematics, Computer Science and Artificial Intelligence.