seminars
Rafał Kucharski and his research group at Jagiellonian University from the inside. Stay up to date with the latest achievements, meet researchers we are working with.
-
Meeting with Fryderyk Falniowski
Fryderyk Falniowski presented his work titled Heterogeneity, Reinforcement Learning and Chaos in Population Games. Abstract: Inspired by the challenges at the intersection of Evolutionary Game Theory and Machine Learning, we investigate a class of discrete-time multi-agent reinforcement learning (MARL) dynamics in population/non-atomic congestion games, where agents have diverse beliefs and learn at different rates. These congestion games, a well-studied class of potential games, are characterized by individual agents having negligible effects on system performance, strongly aligned incentives, and well-understood advantageous properties of Nash equilibria. Despite the presence of static Nash equilibria, we demonstrate that MARL dynamics with heterogeneous learning rates can deviate from these equilibria, exhibiting instability and even chaotic behavior and resulting in increased social costs. Remarkably, even within these chaotic regimes, we show that the time-averaged macroscopic behavior converges to exact Nash equilibria, thus linking the microscopic dynamic complexity with traditional equilibrium concepts. By employing dynamical systems techniques, we analyze the interaction between individual-level adaptation and population-level outcomes, paving the way for studying heterogeneous learning dynamics in discrete time across more complex game scenarios. -
Seminar on communication in MARL
Anastasia presented and overviewed the broad field of Communication in MARL. -
Meeting with Paweł Gora
Hosting Paweł Gora who presented his works, including traffic lights optimization in large scale simulator that he has implemented. -
Meeting with Zafeira Gkioulou
Meeting with Zafeira Gkioulou and discussion on the future of LajkBus. -
Meeting with prof. Vitalii Naumov
Hosting prof. Vitalii Naumov and talking about Optimizing public transport timetables with Genetic Algorithms. -
About potential applications of Wardropian Cycles
Michał Bujak and Michał Hoffmann presented their thoughts on potential applications of Wardropian Cycles in our research.