Fundamentals of Reinforcement Learning : The K-bandit Problem, Illustrated

Welcome to GradientCrescent's special series on reinforcement learning. This series will serve to introduce some of the fundamental concepts in reinforcement learning using digestible examples, primarily obtained from the" Reinforcement Learning" text by Sutton et. Note that code in this series will be kept to a minimum- readers interested in implementations are directed to the official course, or our Github. The secondary purpose of this series is to reinforce (pun intended) my own learning in the field. Reinforcement learning has quickly captured the imagination of the general public, with organisations such as Deepming achieving success in games such as Go, Starcraft, and Quake III, along with more practical achievements such as disease detection and self-mapping.