Are real analysis and measure theory essential to learn for data science? [closed]

Question

Some people say data scientists don't necessarily need to know real analysis and measure theory, but for others, real analysis and measure theory are very important for the undersdanding of kernel methods, stochastic processes etc. Should I learn real analysis and measure theory? I don't want to be too academic

Answer 1

Neither of real analysis or measure theory are necessary for data science. Most of the mathematics you need to know is at the undergraduate level and these courses should suffice:

1. Calculus
2. Linear Algebra
3. Probability and Statistics
4. Optimization

Most of machine learning involved in data science (that deals with application vs creating novel machine learning algorithms), requires a linear combination of concepts learnt in the above mathematics courses. For example, you can understand enough kernel methods well enough if you understood linear algebra (inner products, matrices etc.).

Also, data science has to do with practice and experimentation. A lot of key concepts become clearer only through experimentation and if you need to understand a new method, the best way is to read up specifically about it and then experiment with it. In this respect, I feel studying an entire branch of involved mathematics is not worth it if your goal is to understand a machine learning method -- you can always ask a Guru in such situations to explain what you're missing at a conceptual level :-)

Answer 2

No, you don't need to understand measure theory and real analysis to do machine learning in data science. However, it'd be hard to for you to read academic papers (eg: kernel methods) if you don't have the knowledge.

Unless you want to be a mathematician or wish to pursue a Phd, you really don't need to know too much about the theories. In fact, most of the engineers working in the field I know don't understand statistics at all. They simply use some framework, get some good results, then get paid.