Computer Science > Computer Science and Game Theory
[Submitted on 3 Feb 2024]
Title:Team Collaboration vs Competition: New Fictitious Play Dynamics for Multi-team Zero-Sum Games
View PDFAbstract:This paper presents a new variant of fictitious play (FP) called team-fictitious-play (Team-FP) that can reach equilibrium in multi-team competition, different from the other variants of FP. We specifically focus on zero-sum potential team games with network separable interactions (ZSPTGs), unifying potential games (if there is a single team) and zero-sum polymatrix games (if each team has a single member) due to their wide range of applications from robotics to financial markets beyond two-team games. Similar to the FP dynamics, in Team-FP, agents follow a simple behavioral rule where they respond (with some inertia and exploration in the update of actions) to the last actions of the neighboring team members and the beliefs formed about the other neighbors' strategies as if the opponents are playing according to some stationary strategy. We show the almost sure convergence of the empirical averages of teams' action profiles to near team-Nash equilibrium in ZSPTGs under standard assumptions on the step sizes used. We formulate a bound on the approximation error, decaying with the exploration in the agents' responses. We further examine the performance of the Team-FP dynamics numerically.
Submission history
From: Muhammed Omer Sayin [view email][v1] Sat, 3 Feb 2024 13:32:12 UTC (694 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.