I would go with probability theory but try to probe your Instructor through the course to see if they can give you more insight/advice on how these concepts are

I would go with probability theory but try to probe your Instructor through the course to see if they can give you more insight/advice on how these concepts are translated into ML. You are right - outside of classroom setting, it is quiet difficult to learn maths on your own and yes, maths (especially probability and stats) are precursor to become good at ML and Data Science.