Imitation Learning with Concurrent Actions in 3D Games

In this work we describe a novel deep reinforcement learning neural network architecture that allows multiple actions to be selected at every time-step. Multi-action policies allows complex behaviors to be learnt that are otherwise hard to achieve when using single action selection techniques. This work describes an algorithm that uses both imitation learning (IL) and temporal difference (TD) reinforcement learning (RL) to provide a 4x improvement in training time and 2.5x improvement in performance over single action selection TD RL. We demonstrate the capabilities of this network using a complex in-house 3D game. Mimicking the behavior of the expert teacher significantly improves world state exploration and allows the agents vision system to be trained more rapidly than TD RL alone. This initial training technique kick-starts TD learning and the agent quickly learns to surpass the capabilities of the expert.