I think Math for ML is __fantastic__. And based on the curriculum Jason has published for ML, it seems very __very__ promising. I've done a fair bit of ML @ Cornell, so I've had exposure to a lot of the material he plans on covering. However, I glossed over a lot of the theory because some weakness in my math. I feel like this has been remediated with M4ML and ML should expand and solidify my understanding.

Edit: Wrt to computer graphics, going through M4ML and the ML sequence will really help you understand what's happening there. Convolutional Neural Nets, Gaussian Splatting, all rely on these same principles.