Stochastic Games on Large Graphs in Mean Field and Sparse Regimes

Explore stochastic games on large graphs in mean field and sparse regimes in this SIAM Activity Group on Financial Mathematics and Engineering virtual talk. Delve into the theory of mean field games and its limitations in modeling stochastic dynamic games with many interacting players. Examine recent efforts to overcome these limitations by incorporating heterogeneous interactions, particularly those governed by networks. Focus on a case study of linear-quadratic stochastic differential games where players are labeled by graph vertices and interact symmetrically with nearest neighbors. Investigate large-scale asymptotics for various graph sequences, emphasizing the differences between sparse and dense regimes. Learn about the basic model setup, mean field games and control, large graphs, approximate equilibrium, and automorphisms. Gain insights into this complex topic from speaker Daniel Lacker of Columbia University in this hour-long presentation.