In supervised learning each data example has a label.
Imitation learning is mapping from observations to actions and is generally considered part of reinforcement learning.
The primary difference between supervised learning and reinforcement learning are when the labels / rewards are available. Reinforcement learning labels / rewards are often time delayed. In supervised learning, the labels / rewards have to be available concurrently with all the data. In reinforcement learning, the agent might make several actions before receiving a reward.