Dueling Network Architectures for Deep Reinforcement Learning

Ziyu Wang, Tom Schaul, Matteo Hessel +3

In recent years there have been many successes of using deep representations in reinforcement learning. Still, many of these applications use conventional architectures, such as co…