When it comes to “what all we need to learn in terms of Mathematics for machine learning” and specially for people with not much background in either data Science or not Major in Mathematics, it becomes little challenging.
There are tons of videos on YouTube and searchable material on google and this is an attempt to help out a beginner with some pointers to get started.
Personally i like this line when it comes to machine learning:
“Machine learning is a field of study that gives computers the ability to learn, without being explicitly programmed“
Kind of three areas of Mathematics that i think are useful for Machine Learning are :
- Linear Algebra
- Probability
- Calculus
If I need to map what area of Mathematics can be useful in what area of Machine learning then at a very high level it looks like this :
Mathematics | Machine Learning |
Linear algebra – Vectors and Linear spaces | How we represent data |
Linear algebra – Matrix theory | Transformation of data |
Probability | Encapsulate fluctuation in data |
Calculus | Optimize to models for best fit |
Here are some helpful videos specially for the beginner level:
Very good read if you are wondering how much Math u need to learn for ML
Create an Amazon AWS account and loging to http://aws.training ( It’s free )
Here got through these two courses:
- Math for Machine Learning
- Linear and Logistic Regression
How a function is defined in Mathematics. ML models eventually come up with a function 😉
Linear combinations of vectors
Using Matrices to solve equations
Good and long video on how mathematics is used in ML
What is overfitting and how to use regularization to mitigate it
Basics around logistic regression and classification problems
Machine Learning Videos for Beginners
Have fun ! and feel free to recommend some other good resources for this topic.