Philip Jordan

Philip Jordan is a Doctoral Assistant at École Polytechnique Fédérale de Lausanne (EPFL) in the School of Engineering, Institute of Mechanical Engineering (IGM), working within the SYCAMORE laboratory. His research focuses on multi-agent systems and optimization problems.

He received both his Bachelor's and Master's degrees in Computer Science from ETH Zurich, with a semester abroad at Princeton University during his undergraduate studies. His Master's thesis investigated independent learning in Markov potential games.

His primary research interests include multi-agent reinforcement learning, optimization algorithms, game theory, and their applications in distributed systems. His work often addresses challenges in constrained environments and decentralized decision-making processes where multiple agents interact with shared constraints.

Philip has published in top-tier conferences including AISTATS and AAMAS in 2024, with research focusing on independent learning algorithms for constrained Markov potential games and decentralized federated policy gradient methods with Byzantine fault tolerance. His publications demonstrate strong theoretical foundations coupled with practical implementations.

He currently serves as a Doctoral Assistant in the SYCAMORE laboratory at EPFL, where he conducts research in systems control and multiagent optimization. His work combines theoretical analysis with practical implementation, as evidenced by the publicly available code repository for his AISTATS 2024 publication.

Prior to his doctoral studies, Philip co-founded a start-up in the field of smart contracts, demonstrating practical application of his theoretical knowledge in blockchain technology.