OpenAI Spinning Up

Very helpful for understanding the whole Reinforcement Learning structure. Especially for me because I've already tried my first implementation of a common solution in RL, Q-Learning algorithm with ε-greedy policy. This documentation helps me mostly to understand RL formally and mathematically with its organized and self-consistent descriptions, and to learn about its terminology, which I desperately need for finding a better RL model for my thesis.

Different and better than other RL introductions I have read online, is that this documentation does not start with Markov Decision Process. Although I've learned the concept of MDP in lecture "signals and systems", it seems bit outdated to model RL based on MDP, because nowadays algorithms are already very common, and the concepts in RL are very easy to accept.