How to learn certain Maths to understand machine Learning papers?

I have done the deeplearning.ai course on deep learning. But I cannot Understand equations like

minGmaxDV(D,G)=Ex∼pdata(x)[logD(x)]+Ez∼pz(z)[log(1−D(G(z)))]

What kind of Maths am I am supposed to learn? I know Calculus basic Multivariable Calculus and Linear Algebra. So should I learn Probability (Advance probability maybe)?

• it would help if you could tell us which part you don't understand – oW_ Jul 14 at 22:23
• I should you should instead focus on learning the notations rather than concepts. – Shubham Panchal Jul 15 at 0:03
• I am confused about Expectation values? – harshtiwari Jul 15 at 8:19

Mathematical expectations are a concept from probability theory. Expressions like $$E_x[...]$$ denote conditional expectations and I think you will have trouble understanding Machine Learning concepts without understanding conditional expectations. (Understanding them as a mathematical concept, not on an intuitive level).

My recommendation is that you look for a beginner's course in probability theory to get a grasp of the following concepts:

• Random variables
• Probability distributions
• Stochastic independence
• Mathematical expectation
• Conditional expectation/conditional probability
• Law of total probability
• Bayes' theorem

There are probably courses designed specifically for people interested in Machine Learning. Have fun!

As Elias mentioned, the expectations are related to random variables and you would be good to go if you know about conditional probability, multivariate probability, joint and marginal distributions. I would suggest you take a course that has a syllabus on the lines of https://secure.oregonstate.edu/ap/cps/documents/view/134169.

• +1 on multivariate probability, joint distributions and marginal distributions! – Elias Strehle Jul 16 at 16:39