The math can be pretty scary in RL. Researchers use them to scare people. But take my words, it can be understood intuitively. Just hope the process is not that painful for you. It was painful for me when I started it a few years ago. But glad to hear your feedback.
But for this particular article, it is not important to understand the equations. The main idea in imitation learning is simply using supervised learning to imitate the experts. Then due with those drifting problems. The equations are mostly from the Policy Gradient. This article will introduce you to some of the mathematical terms if you are interested.
RL — Policy Gradient Explained
Policy Gradient Methods (PG) are frequently used algorithms in reinforcement learning (RL). The principle is very…
I am trying to put together materials with different difficulty levels. I have not published the introduction one. Stay tuned.