Song Xuanye
Keynote Speaker
Currently Research Fellow at Nanyang Technological University, Singapore. PhD degree from Université Paris-Cité, France, under the supervision of Prof. HuyenPham and Prof. Noufel Frikha.
Challenges for problems in mean-field setting In this presentation, I will show my previous work about how to deal with thechallenges for problems in mean-field setting. One of the biggest challenges lies inthe bad properties of the Wasserstein space. For example, it is infinite dimensionand it is only a Polish metric space, without any computation properties as Hilbertspace. I will present how the challenges can be overcome in some specificproblems. One is about solving mean-field control problem in continuous time viareinforcement learning method. Another is about numerically solving mean-fieldMcKean-Vlasov SDE via Euler-Maruyama Scheme. I will show the theoretical resultswe obtained. At the end, I show the future work.