(1)
Tavakoli, A.; Pardo, F.; Kormushev, P. Action Branching Architectures for Deep Reinforcement Learning. AAAI 2018, 32.