Multi-Agent Meeting Scheduling: A Negotiation Perspective

Bram M. Renting, Holger Hoos, Catholijn M Jonker

Keywords: Multi-Agent Systems, Negotiation, Meeting Scheduling

Abstract:

In multi-agent systems (MAS), applications that directly interface with daily human activities represent a rich avenue for exploration. This paper dives into a potentially impactful application of MAS, targeting a well-known real-world challenge: meeting scheduling. While there have been previous efforts to address this challenge, we believe that the time is right to revisit this task as a blue-sky challenge for the MAS community.

Traditional scheduling methodologies rely on static, sub-optimal support tools that are susceptible to inefficiencies that include repeated rescheduling, and the overhead for the humans affected per scheduling attempt remains substantial. This opens an intriguing challenge for the MAS community: What if a collection of autonomous agents could extend human capabilities, designed to adapt and negotiate, making scheduling more dynamic and less time-consuming? The potential of collective time saved is substantial, not only in a reduction of human effort due to fewer rescheduling attempts, but also in better alignment of schedules. Furthermore, the privacy of participants can be better preserved.

We argue that the richness of this domain is of interest to the MAS community and that recent advances in AI open up new ways for tackling this challenge. In this paper, we set the stage for this research direction, focussed on the use of MAS to support an age-old, yet fundamental and pervasive task.

Type Of Paper: Work-in-progress paper (max page 6)

Anonymous Submission: Anonymized submission.

Submission Number: 23