Doctoral School in Mathematics
Convergence and optimality of neural networks for reinforcement learning