This is a good write up to build an intuition around Reinforcement Learning, but I don’t know if it qualifies as a “Real world problem”. This is actually a simpler problem than the Atari games I usually see in RL articles. I still have yet to see a good article on how a business used RL to optimize a process in real time (outside of self driving cars). Any that you know of? We have proposed a project using RL for logistics dispatch optimization, but couldn’t convince the client to take the risk.