no code implementations • 27 Mar 2024 • Sharif Azem, David Scheunert, Mengguang Li, Jonas Gehrunger, Kai Cui, Christian Hochberger, Heinz Koeppl
The advent of unmanned aerial vehicles (UAVs) has improved a variety of fields by providing a versatile, cost-effective and accessible platform for implementing state-of-the-art algorithms.
no code implementations • 23 Jan 2024 • Christian Fabian, Kai Cui, Heinz Koeppl
This hybrid graphex learning approach leverages that the system mainly consists of a highly connected core and a sparse periphery.
no code implementations • 20 Dec 2023 • Anam Tahir, Kai Cui, Heinz Koeppl
Empirically, the proposed methodology performs well on several realistic and scalable wireless network topologies as compared to a number of well-known load balancing heuristics and existing scalable multi-agent reinforcement learning methods.
no code implementations • 20 Dec 2023 • Anam Tahir, Kai Cui, Bastian Alt, Amr Rizk, Heinz Koeppl
In this work, we devise a decentralized AoI-minimizing transmission policy for a number of sensor agents sharing capacity-limited, non-FIFO duplex channels that introduce random delays in communication with a common receiver.
1 code implementation • 17 Dec 2023 • Kai Cui, Gökçe Dayanıklı, Mathieu Laurière, Matthieu Geist, Olivier Pietquin, Heinz Koeppl
We propose a novel discrete time version of major-minor MFGs (M3FGs), along with a learning algorithm based on fictitious play and partitioning the probability simplex.
no code implementations • 12 Jul 2023 • Kai Cui, Sascha Hauck, Christian Fabian, Heinz Koeppl
However, multi-agent RL (MARL) remains a challenge in terms of decentralization, partial observability and scalability to many agents.
no code implementations • 19 Mar 2023 • Kai Cui, Christian Fabian, Heinz Koeppl
In this work, we propose a novel discrete-time generalization of Markov decision processes and MFC to both many minor agents and potentially complex major agents -- major-minor mean field control (M3FC).
no code implementations • 15 Sep 2022 • Kai Cui, Mengguang Li, Christian Fabian, Heinz Koeppl
Thus, we combine collision avoidance and learning of mean-field control into a unified framework for tractably designing intelligent robotic swarm behavior.
no code implementations • 8 Sep 2022 • Kai Cui, Anam Tahir, Gizem Ekinci, Ahmed Elshamanhory, Yannick Eich, Mengguang Li, Heinz Koeppl
The analysis and control of large-population systems is of great interest to diverse areas of research and engineering, ranging from epidemiology over robotic swarms to economics and finance.
1 code implementation • 8 Sep 2022 • Christian Fabian, Kai Cui, Heinz Koeppl
Although the field of multi-agent reinforcement learning (MARL) has made considerable progress in the last years, solving systems with a large number of agents remains a hard challenge.
no code implementations • 8 Sep 2022 • Christian Fabian, Kai Cui, Heinz Koeppl
Graphon mean field games (GMFGs) on the other hand provide a scalable and mathematically well-founded approach to learning problems that involve a large number of connected agents.
1 code implementation • 9 Aug 2022 • Anam Tahir, Kai Cui, Heinz Koeppl
In this work, we consider a multi-agent load balancing system, with delayed information, consisting of many clients (load balancers) and many parallel queues.
1 code implementation • 14 Jun 2022 • Yingguang Yang, Renyu Yang, Yangyang Li, Kai Cui, Zhiqin Yang, Yue Wang, Jie Xu, Haiyong Xie
More specifically, we consider the social bot detection problem as a user-centric subgraph embedding and classification task.
no code implementations • 30 Mar 2022 • Kai Cui, Wasiur R. KhudaBukhsh, Heinz Koeppl
We propose an approach to modelling large-scale multi-agent dynamical systems allowing interactions among more than just pairs of agents using the theory of mean field games and the notion of hypergraphons, which are obtained as limits of large hypergraphs.
1 code implementation • ICLR 2022 • Kai Cui, Heinz Koeppl
Recent advances at the intersection of dense large graph limits and mean field games have begun to enable the scalable analysis of a broad class of dynamical sequential games with large numbers of agents.
no code implementations • 30 Apr 2021 • Ramzi Ourari, Kai Cui, Ahmed Elshamanhory, Heinz Koeppl
Collision avoidance algorithms are of central interest to many drone applications.
no code implementations • 30 Apr 2021 • Kai Cui, Anam Tahir, Mark Sinzger, Heinz Koeppl
Multi-agent reinforcement learning methods have shown remarkable potential in solving complex multi-agent problems but mostly lack theoretical guarantees.
Multi-agent Reinforcement Learning reinforcement-learning +2
no code implementations • 2 Feb 2021 • Kai Cui, Heinz Koeppl
We show that all discrete-time finite MFGs with non-constant fixed point operators fail to be contractive as typically assumed in existing MFG literature, barring convergence via fixed point iteration.
1 code implementation • 7 Oct 2018 • Kai Cui, Zhi Jin, Eckehard Steinbach
Color demosaicking (CDM) is a critical first step for the acquisition of high-quality RGB images with single chip cameras.