This document provides an introduction to reinforcement learning. It discusses how recent research has used deep neural networks (DNNs) and convolutional neural networks (CNNs) to approximate Q-functions and solve problems. This research has achieved superhuman performance on tasks like playing Atari 2600 games and Go, and has also been applied to 3D games and robot control. The document then provides overviews of machine learning, supervised learning, unsupervised learning, and reinforcement learning, and provides examples of each type.