What Should I Read to Startup with AI and ML?
I’m listing out all the prerequisites you’ll need before starting ML or AI. Some of the prerequisites I’ve listed, you might find basic or even stupid since you’d already know them but perhaps it would be useful to somebody else.
- Matrix Multiplications: You need to have a clear understanding of how these work. Since most of the ML courses you’ll take or the books you might read, might not cover this as it school level so I’d say a refresher to this before you start would be great.
- Statistics: Again, a high school level prerequisite, but a refresher of concepts like mean, variance, standard deviation, types of distribution especially Gaussian, would do you good before you begin. If you have more time to dedicate to this, also read up on measures of similarity and distances like the Euclidean distance and the cosine similarity. However, most books or courses will cover this part albeit not in too much detail.
- Probability: Important. Brush through the basics and read up a bit on Bayes’ theorem before you begin since the theorem forms the backbone of the Naive Bayes ML Model you might come across.
- Calculus: Here is where it gets a little tricky. If you do not know calculus at all, I recommend take a calculus class before entering ML since you might be able to get by at the introductory level without exquisite calculus knowledge but further in the domains that lead up like Deep Learning or Reinforcement Learning, calculus is a basic must.
- Linear Algebra: At an introductory level, you might come across daunting terms like the Principal Component Analysis or Dimensionality Reduction Techniques, and for you to understand this fully you will need to know about Eigen Vectors and Eigen Values. Just a small crash on them would do you great and you won’t be overwhelmed by the math when it is actually thrown at you.
- Programming Language: This stands out as a bit stupid, but if you are going to be doing ML, you will need to code and while you might get by with no previous coding experience if you take up an online course because they do tend you teach you and give you some practice, it might not be the same with books. I’d recommend Python or MATLAB.
’ve listed down all the prerequisites that I needed when I was about to enter ML. There might be more and I’m sure people will point that out but I hadn’t needed anything other than these to begin.
I might be a little out of line here, but before I can put links to the books I would highly recommend you take a free online course rather than a starting with a book. It is way better an experience than a book trust me because I’m a guy who would prefer a book any day. Besides with an online course, you get the required coding hands-on too by completing graded assignments.
Here’s the course all of the internets would recommend and rightly so:
or you can GO for Youtube videos like