In Reinforcement learning or deep reinforcement learning, the user gets positive as a reward and negative feedback. Reinforcement learning is a type of Machine learning that constructs the agents to take actions according to the environment. In reinforcement learning, the agent will be awarded a point for success and a negative point as feedback for failure. He agent learns how to react according to the environment from both reward and feedback.
For example, consider a batsman trying to hit the ball for six. He swings the bat and if he misses, he gets a negative and gets a positive point if he hits. So, the batsman learns from these positive and negative points and learns which ball to hit for six.
You can watch this video on Machine learning Full course to have an overview of reinforcement and other types of Machine learning: