Do we need maths for Machine Learning?

Many readers and beginners have asked me multiple times whether we really need maths for machine learning, and I believe this is the most often asked question by many of my friends and students, so I am writing an article in which I will answer this topic for once last time.

In this article, I will discuss if we need math for machine learning, and if so, what topics are important and their use cases.

Do we need Maths for Machine learning?

First and first, let me clarify the myth that machine learning is meaningless without math.
If you’re a beginner looking to work in industry or company, math isn’t the most important qualification for machine learning. That most likely contradicts what you’ve heard in the past.

Topics of maths used

Here I have discussed the topics of maths we need with their use cases.

Linear algebra

Linear Algebra helps in the improvement of graphics processing in Machine Learning. And it helps in the systematic visualisation of data as well as determining which model would perform best on model for example, if we have growing linear data as seen in the image below. so we can see a linear model be a best fit for this type of dataset.

Here we are using linear regression model. we have many models like linear regression for which algebra is important.


Statistics is an important part of machine learning. We’ll need some fundamental statistics formulae like mean, median, standard deviation, and so forth. These are utilised to obtain more valuable data information. It also helps you in drawing useful insights from raw data.


Probability is the capacity to forecast the likelihood of future events. and will mostly certainly not be used in most circumstances. Naive bayes is a common probability and statistics-based model.

This is the fundamental mathematics required for machine learning. You’ve probably heard of calculus, which is used in math. Yes, it is used, but if you are a beginner, you do not require it.How much maths do we do?

Where do I need maths usually

After spending more than 2 years on machine learning I don’t use maths very often.

  1. My main use cases for math are when I’m learning about a new model and the algorithm behind it. I must learn the formula and algorithm, although in most cases, this is not necessary.
  2. My second use case is when I put my data on a graph to determine which model is best suited for the dataset.
  3. The third situation is when I need to know my machine learning model’s accuracy, for example, here I use mean absolute error, mean squared error, and stats to get to know my model better and gain more insights from the data.
  4. I occasionally utilise statistics to get to know my model better and gain new insights from the data.

Well, if you like this article you can check out my articles for more interesting articles in the field of artificial intelligence and machine learning.

If you like my article and efforts towards the community, you may support and encourage me, by simply buying coffee for me


If you found this article useful please appreciate it by giving claps and follow me for more interesting articles. Well, I have good news for you I would be bringing more articles to explain machine learning concepts and models with codes so leave a comment and tell me how excited are you about this.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Aviral Bhardwaj

Aviral Bhardwaj

One of the youngest writer and mentor on AI-ML & Technology.