A Reinforcement Learning demo using a simple gridworld agent. Shows concepts like reward, Q-function, and Bellman equation without using neural networks. - View it on GitHub