Fetching the paper…

SBEED: Convergent Reinforcement Learning with Nonlinear Function Approximation · Around