One of three basic machine learning paradigms (alongside supervised learning and unsupervised learning). Reinforcement learning is about taking suitable action to maximize reward in a particular situation. In this family of algorithms the input is an initial state. There are many possible outputs as there variety of solutions to a particular problem. The training phase is based upon the input - the model returns a state and the user will decide to reward or punish the model based on its output. The best solution is decided based on the maximum reward.
Next: Jupyter, JupyterLab