This post discusses the recent exciting advancements in Artificial Intelligence (AI) such as AlphaGo defeating human players and OpenAI’s bot beating professional gamers in DOTA2. It introduces Reinforcement Learning (RL) as the magic behind these algorithms, where an agent interacts with an environment to maximize rewards. Key concepts like models, policy, and value functions are explained, leading to discussions on Markov Decision Processes (MDPs) and Bellman equations. The post then delves into common approaches for solving RL problems like Dynamic Programming and Policy Iteration.Overall, the content provides a comprehensive overview of RL concepts and algorithms, making it a valuable resource for beginners in the field.
https://lilianweng.github.io/posts/2018-02-19-rl-overview/