Reinforcement learning is based on the concept of learning from interaction with a surrounding environment and consequently rewarding positive actions taken in that environment. In reinforcement learning, we refer to our algorithm as the agent because it takes action on the world around it:
When an agent takes an action, it receives a reward or penalty depending on whether it took the correct action or not. Our goal in reinforcement learning is to let the agent learn to take actions that maximize the rewards it receives from its environment. These concepts are not at all new; in fact, they've been around for quite some time. What has allowed reinforcement learning to achieve such great heights has been the combination of new advances in deep learning, coupled with the computing power to handle increasingly complex scenarios.
There are...
