Background

Deep reinforcement learning (DRL) methods such as the Deep Q-Network (DQN) have achieved state-of-the-art results in a variety of challenging, high-dimensional domains. This success is mainly attributed to the power of deep neural networks to learn rich domain representations for approximating the value function or policy.

While current methods are able to achieve outstanding results, the current algorithms aren't sample efficient and are still unable to beat the top human scores.

We believe that by taking imitation learning a level further, we can not only learn complex problems in a fast and efficient manner but also reach high scores which are far beyond the abilities of current algorithms.