Call for Papers

Special Issue Multi-agent Dynamic Decision Making and Learning

Guest editors: 
• Konstantin Avrachenkov (INRIA Sophia Antipolis),  k.avrachenkov@inria.fr
• Vivek S. Borkar (IIT Bombay), borkar.vs@gmail.com
• U. Jayakrishnan Nair (IIT Bombay), jayakrishnan.nair@ee.iitb.ac.in

As a large and ever-increasing part of our economic and social interactions move to the cyberspace, data-driven algorithmic decision making by autonomous agents is fast becoming an integral and inseparable part of our lives. These agents are competing in uncertain and volatile environments and must in turn learn aspects thereof, and of each other, in order to dynamically optimize their performance. What’s more, even the humans in the loop are obliged to depend more and more on data-driven signals for their own decision making, e.g., on automated rankings and recommendations. Given the inherently distributed, strategic, dynamic nature of this ethos, learning in dynamic games, with its broad spectrum of modeling and analysis tools, is a prime candidate for providing this endeavour the theoretical underpinnings, with a balance between unification of the mathematical substructure while retaining the distinct flavors and diversity of the competing paradigms. On modeling front, this ranges from dynamic cooperative games to mean field and evolutionary games, and for learning paradigms, from reinforcement learning to learning by imitation.

This nascent role of dynamic games has already registered its presence in many different ways and is increasingly doing so. The time is thus ripe for taking stock of where we are and where we should be heading. This is the motivation behind this special issue. The subarea is still too young to be put into a straitjacket of well defined boundaries. So what we have here is a list of tentative themes, to be interpreted suitably with the proviso that due attention is given to the key words ‘dynamic’ and ‘games’ that are the signature of the journal, and ‘multi-agent’ and ‘learning’, that narrows the theme down to the aforementioned issues. Submitting authors should keep in mind that all of these should be reflected in the content, though not necessarily in equal measure, and missing out on any of these may be considered an adequate reason for unsuitability for this issue. Also to be kept in mind is the intended flavor, i.e., that of algorithmic or data-assisted learning by humans and autonomous software agents. Most importantly, we seek articles with serious theoretical contributions. Articles based solely on data analysis or numerical experiments will be turned down.

Those desirous of submitting an article for the special issue are encouraged to send the guest editors an email indicating their intent, to help them in planning.

Topics of interest: 
• Game-theoretic issues in multiarmed bandits
• Reinforcement learning in non-cooperative games
• Distributed learning algorithms of game-theoretic flavour
• Learning in cooperative games
• Learning models in mean field games
• Dynamic aspects of adversarial machine learning, GANs
• Evolutionary paradigms in game-theoretic learning 

Submission timetable:
The final deadline is June 30, 2021. Papers that still require major revision after the second round will not be accepted for the special issue and will be treated as submissions to a regular issue. We encourage early submissions and the submissions will be processed as soon as they are received. The accepted papers will appear online in advance of the production of the full special issue. 

Click to:
- download the Call for Papers
- see the submission guidelines 
- submit your manuscript