Deriving Backpropagation from Scratch

Backpropagation is the backbone of training deep neural networks. In this post, we will unravel the mathematics behind this critical algorithm, starting with the basics and building up to the full derivation.

The Basic Idea

The goal of training a neural network …

The Synonyms of Intelligence: Compression, Abstraction, and Learning

Imagine you’re given two tasks.

Task 1: Memorize the multiplication tables up to 12x12. This means memorizing 144 results, each a combination of two numbers and their product.

Task 2: Learn the rules of multiplication.

At …

Step 1: Import NumPy and Prepare Data

In any machine learning or data science project, data preparation is a critical step. It ensures that the data you’re working with is in the right format, and that you have all the necessary tools at your disposal to manipulate it. In this step, …