On the Mathematics of Machine Learning

Throughout university I have struggled and still struggle with Imposter Syndrome and I am sure everyone else has had similar feelings at one point or another. One of the main areas this occurs most often for me is mathematics. I know I am not bad at math, throughout my primary and secondary education I was in advanced maths for my year level. I know when to use derivatives and when to integrate, I understand summation notation, and I can handle basic math in my head. However I don’t feel like I fully grasp math or that it intuitively “clicks” for me like it would like. So I have decided to take action…

There are many ways to learn math: textbooks, videos, and online courses. With the many different sources it can be intimidating to know where to start. I decided to use the Mathematics For Machine Learning (which is available for free online) to create a guide for what maths to learn in what order. The MML book has a clear pathway which makes knowing the order in which to learn the maths.

I will be starting with the Mathematical foundations and follow the following order:

• Linear Algebra
• Analytic Geometry
• Matrix Decompositions
• Vector Calculus
• Probability and Distribution
• Continuous Optimisation

I will be posting sporadically as I work my way through each of these categories with what I have learned and the resources that I have used, some sections may have several posts and others less, only time will tell.

Now let’s do some math…