Thread by @PrasoonPratham

What you see below is called "Sigma notation".

Something I've encountered frequently in the math for machine learning and number theory.

Here's how it works.
🧵 👇🏻

This weirdly shaped 'E' is the capital greek alphabet called 'Sigma' and it stands for summation.

Essentially what it allows us to do is represent equations where we add multiple items, let me show you what I mean.

Let's say we have to add squares of numbers from 1 to 10, you'd write it as an equation like this.

(The square of a number is basically multiplying the number by itself, like 2², read as 2 squared = 2 x 2 =4)

You can write the same equation using Sigma notation.

Doesn't it look cleaner and shorter?

I like to think of Sigma as a 3 step process (like a 'for' loop if you're a programmer), you start from the bottom and then go towards the top.

In this case:

1. You start with i=1 (on the bottom)
2. Square it
3. Add 1 to the previous value of i
4. Square it
5. Add this to the previous squared number
6. Repeat until i reaches the number on the top which is 10 here.

At times you will notice a number in front of sigma, this basically multiplies with the entire equation with that particular number.

I first encountered sigma in the RSME (root mean squared error) formula, a commonly used metric while training machine learning models.

Before we bring this thread to end, try solving this sigma equation.

If you like bite-sized content on math, machine learning and programming consider following me.

Until next time, peace out.