Math
If you want to do research, then you need a very strong basis in statistics and mathematics. Basically, Maths / CS degree level Calculus, Statistics & Probability, and Linear Algebra.
If you want to be a practitioner, you might be able to take some shortcuts:
I actually found the applied math section in the free online Deep Learning book to be focused, and clear. Andrew Ng's Machine Learning | Coursera course also has some basic relevant maths reviews.
If you want a bit more in-depth background and practice, you can either:
Take calculus 1, linear algebra 1, and statistics and probability undergrad courses for physics / chemistry / biology / statistics students.
Take the equivalent online MOOCs. There are many options, here are some good ones IMHO:
Bayesian Methods for Machine Learning | Coursera
Calculus: Single Variable Part 2 - Differentiation | Coursera
Mathematics for Machine Learning: Linear Algebra | Coursera
Statistics with R | Coursera
Calculus One | Coursera
Programming
As for CS background, you don't need much to start with, beyond basic Python programming skills. Though, if you do want to have a career in the space (I assume you are looking for a practitioner career, not research), I would recommend you strengthen up your foundations by taking some courses in:
Programming - such as Introduction to Computer Science and Programming Using Python,
Data structures
Algorithms - such as Algorithms: Design and Analysis
Data science - like Data Science
There are many good sources for these across MIT OpenCourseWare, Harvard EdX, Stanford online, Coursera, etc. Choose the ones you like best.
ML and DL
Finally, to get started with ML, check out this post: A beginners guide to learning Deep Learning – Good Audience. It will take quite a while to go through all of it, but even halfway through you could already start practicing and writing your own somewhat-toy projects. I think that getting to do that is the most important step.
See this Machine Learning Course for better knowledge :
Back