Prof. Zhang is very passionate about Reinforcement Learning and has a wealth of knowledge. He isn't always the best at gauging student understanding as he often rushes ahead before everyone has adequate understanding. Your entire grade is just reproducing a figure from the textbook every week which he has reference solutions to on his GitHub (which he is very open about). It's very easy to get a good grade in this class by just copying the solutions but I would advise against doing this blindly. A lot of the joy of taking this class is understanding the algorithms and taking the time to soak it all in. Conceptually, RL can be hard to follow as there are tons of symbols and it's easy to forget what they all mean. Taking the time to read the textbook can often make lectures a lot clearer.