Assisted Reinforcement Learning is an extension to Reinforcement Learning. ARL allows people and other agents to work with an agent during its learning, critiquing its past actions, informing future actions, and providing additional information about the problem. The aim of Assisted Reinforcement Learning is to create agents that can be taught similarly to how to teach pets and children, with simple dialogue and advice.

This presentation is aimed a those that have little knowledge of reinforcement learning. It does not go into the maths and implementations of ARL or RL. It is assumed that the audience has little to no knowledge of RL.