Games, Decisions & Networks Research Lab

Games, Decisions & Networks Research Lab is led by Dr. Muhammed Omer Sayin at the Department of Electrical and Electronics Engineering, Bilkent University, Turkey. We focus on research challenges in socio-technical systems where humans and technological systems can engage with each other. Our overarching goal is to develop an understanding for the theoretical and algorithmic foundation of learning and autonomy in complex and dynamic systems so that we can solve these challenges with systematic guarantees. We resort to (and bring together) technical concepts and tools from game theory, decision theory and network theory. Application areas include artificial intelligence, intelligent and autonomous systems, robotics, and security.

See the preprint arXiv:2111.11742 and the following poster and the video recording as examples of our recent research projects:

M. O. Sayin, F. Parise and A. Ozdaglar, “Fictitious play in zero-sum stochastic games,” SIAM Journal on Control and Optimization, 2022.
M. O. Sayin, K. Zhang, and A. Ozdaglar, “Fictitious play in Markov games with single controller,” In the Proc. ACM EC, 2022.

We have co-organized the online Games, Decisions & Networks Seminar Series on Fridays at 7 pm Istanbul Time in 2021! You can refer to the seminar webpage for the details of the talks and the links: https://sites.google.com/view/gamesdecisionsnetworks

We are looking for new team members. Please contact us if you are interested in!

News

  • [Feb. 2023] A new preprint is available: M. O. Sayin and O. Unlu, “Efficient-Q learning for stochastic games”, available at arXiv:2302.09806. [url]
  • [Dec. 2022] Our joint work with Professor Chung-Wei Lin’s lab at the National Taiwan University with the title “Reinforcement-learning based job-shop scheduling for intelligent intersection management” will appear in DATE’23.
  • [Aug. 2022] Our paper with the title “On the global convergence of stochastic fictitious play in stochastic games with turn-based controllers” will appear in IEEE CDC’22. [url]
  • [May 2022] A new preprint is available: M. O. Sayin and O. Unlu, “Logit-Q learning in Markov games”, available at arXiv:2205.13266. [url]
  • [May 2022] Our paper with the title “Fictitious play in Markov games with single controller” will appear in ACM EC’22. [url]
  • [Apr. 2022] Our paper: M.O.Sayin, F. Parise, and A. Ozdaglar, “Fictitious play in zero-sum stochastic games” is accepted to SIAM Journal on Control and Optimization. [url]
  • [Apr. 2022] A new preprint is available: M. O. Sayin, “On the global convergence of stochastic fictitious play in stochastic games with turn-based controllers”, submitted to IEEE CDC’22, available at arXiv:2204.01382. [url]
  • [Mar. 2022] Dr. Sayin received the TUBITAK 2232-B International Fellowship for Early-Stage Researchers for his project “Plug-and-Play Decentralized Multi-Agent Reinforcement Learning”
  • [Mar. 2022] Our paper with the title “On the heterogeneity of independent learning dynamics in zero-sum stochastic games” will appear in L4DC’22. [url]
  • [Dec. 2021] A new preprint is available: M. O. Sayin and K. A. Cetiner, “On the heterogeneity of independent learning dynamics in zero-sum stochastic games”, submitted to L4DC’22. [pdf] [url]
Our Poster at NeurIPS’21 [Dec. 2021]
  • [Nov. 2021] A new preprint is available: A. Ozdaglar, M. O. Sayin and K. Zhang, “Independent learning in stochastic games,” ArXiv:2111.11743 [An invited chapter for the International Congress of Mathematicians 2022 (ICM 2022)] [url]
  • [Oct. 2021] Our patent “Managing roadway intersections for vehicles” is granted on Oct. 19, 2021. [url]
  • [Sep. 2021] Our paper “Decentralized Q-learning in Zero-sum Markov Games” will appear in NeurIPS’21. [url]