7
$\begingroup$

I am an Android programmer. Now, I would like to learn machine learning. I know it requires a mathematical background, like statistics, probability, calculus and linear algebra. However, I am a bit lost. Where should I start from? Can someone provide me a road map for how to learn the mathematical background required for machine learning?

| improve this question | | | | |
$\endgroup$
  • 1
    $\begingroup$ Most machine learning algorithms requires very very basic math, like dot product and simple functions like sin, cos (if you know what they are, that's enough)... Then you will encounter my enemy, the Back propagation, which forces you to open Calculous, it is not that complex but for math-dull people like.. everyone in the world, it requires some dedication. Short answer, start by learning ML and lookup math as required. $\endgroup$ – Aus May 9 '17 at 11:11
  • $\begingroup$ Start off by reading Handbook of Neuroevolution through Erlang amazon.com/Handbook-Neuroevolution-Through-Erlang-Gene/dp/…, a great start and requires very basic math. $\endgroup$ – Aus May 9 '17 at 11:15
  • $\begingroup$ A related question: What do I need to study for machine learning?. $\endgroup$ – nbro Jun 2 '19 at 16:28
5
$\begingroup$

You should begin from Dr Andrew Ng machine learning course on Coursera. It's probably the most popular course for newcomers in machine learning. It's a free course.

You should also grab "Elements of Statistical Learning" ebook PDF. It's a free book.

You may want to focus on:

  1. Regression
  2. Cross validation
  3. Bias-variance tradeoff
  4. Decision surface
  5. Gradient descent

And more...

| improve this answer | | | | |
$\endgroup$
  • $\begingroup$ An excellent book, but I'd recommend "An Introduction to Statistical Learning" (by mostly the same authors) first. It's more introductory and is less mathematically demanding. www-bcf.usc.edu/~gareth/ISL There's also a free online course at Stanford by its authors using the book. lagunita.stanford.edu/courses/HumanitiesSciences/StatLearning/… $\endgroup$ – Randy May 22 '17 at 21:00
  • $\begingroup$ If you actually want to learn mathematics for machine learning start with Probability, Linear Algebra, Optimization, Dynamics Systems. Everything in in ML is built out of this stuff. And most people in the field have not really read things properly. $\endgroup$ – mathtick Mar 1 at 15:48
1
$\begingroup$

If you are interested to deepen your statistical concepts before diving into machine learning, i would recommend Introduction to Statistics: Descriptive Statistics course in edX

where you'll learn

  • The fundamental concepts and methods of statistics
  • How to intepret graphical and numerical summaries of data
  • Understand the reasoning behind the calculations, the assumptions under which they are valid, and the correct interpretation of results

The link for course is edX

This will definitely clarify your stat background with added benefit of certification.

| improve this answer | | | | |
$\endgroup$
1
$\begingroup$

Some of the fundamental mathematical concepts required in ML field are as follows:

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

A very recent book availble at Mathematics for Machine Learning covers all these aspects and more.

| improve this answer | | | | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.