Mixing Update Q-value for Deep Reinforcement Learning | IEEE Conference Publication | IEEE Xplore