Long, A.; Blair, A.; Hoof, H. van. Fast and Data Efficient Reinforcement Learning from Pixels via Non-Parametric Value Approximation. AAAI 2022, 36, 7620-7627.