Local Advantage Actor-Critic for Robust Multi-Agent Deep Reinforcement Learning | IEEE Conference Publication | IEEE Xplore