M. Hessel, “Rainbow: Combining Improvements in Deep Reinforcement Learning”, AAAI, vol. 32, no. 1, Apr. 2018.