Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

Zero To Mastery

Build a Simple Neural Network & Learn Backpropagation

via Zero To Mastery

Go to class Write review

Overview

Learn about backpropagation and gradient descent by coding your own simple neural network from scratch in Python - no libraries, just fundamentals. Ideal for aspiring Machine Learning Engineers, Data Scientists, and AI Specialists.
  • Coding neural networks from scratch using only Python
  • What backpropagation is and how it helps machines learn
  • How to break down complicated math into simple, doable steps
  • The easiest way to understand gradients and why they matter
  • What’s really happening when a machine makes predictions
  • How to train a smarter model by adjusting tiny details in code

Syllabus

  •   Introduction
    • Introduction
    • Exercise: Meet Your Classmates and Instructor
    • Course Resources
  •   Neural Networks, Derivatives, Gradients, Chain Rule, and Gradient Descent
    • Introduction to Our Simple Neural Network
    • Why We Use Computational Graphs
    • Conducting the Forward Pass
    • Roadmap to Understanding Backpropagation
    • Derivatives Theory
    • Numerical Example of Derivatives
    • Partial Derivatives
    • Gradients
    • Understanding What Partial Derivatives Dо
    • Introduction to Backpropagation
    • (Optional) Chain Rule
    • Gradient Derivation of Mean Squared Error Loss Function
    • Visualizing the Loss Function and Understanding Gradients
    • Using the Chain Rule to See how w2 Affects the Final Loss
    • Backpropagation of w1
    • Introduction to Gradient Descent Visually
    • Gradient Descent
    • Understanding the Learning Rate (Alpha)
    • Moving in the Opposite Direction of the Gradient
    • Calculating Gradient Descent by Hand
    • Coding our Simple Neural Network Part 1
    • Coding our Simple Neural Network Part 2
    • Coding our Simple Neural Network Part 3
    • Coding our Simple Neural Network Part 4
    • Coding our Simple Neural Network Part 5
  •   Implementing Our Advanced Neural Network by Hand + Python
    • Introduction to Our Complex Neural Network
    • Conducting the Forward Pass
    • Getting Started with Backpropagation
    • Getting the Derivative of the Sigmoid Activation Function(Optional)
    • Implementing Backpropagation with the Chain Rule
    • Understanding How w3 Affects the Final Loss
    • Calculating Gradients for Z1
    • Understanding How w1 and w2 Affect the Loss
    • Implementing Gradient Descent by Hand
    • Coding our Advanced Neural Network Part (Implementing Forward Pass + Loss)
    • Coding our Advanced Neural Network Part 2 (Implement Backpropagation)
    • Coding our Advanced Neural Network Part 3 (Implement Gradient Descent)
    • Coding our Advanced Neural Network Part 4 (Training our Neural Network)
  •   Where To Go From Here?
    • Review This Byte!

Taught by

Patrik Szepesi

Reviews

Start your review of Build a Simple Neural Network & Learn Backpropagation

Go to class

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Sign up for free
Someone learning on their laptop while sitting on the floor.