Stochastic maximum principle for mean-field systems with recursive utilities under model uncertainty

This paper concerns the maximum principle for the stochastic recursive optimal control problem under model uncertainty of mean-field type, that is, the state process and the cost functional depend not only on the control, but also on the law of the state process. Due to model uncertainty, the necessary maximum principle is proved by a weak convergence method and Sion's minimax theorem. Moreover, under additional assumptions, we prove a sufficient maximum principle.