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

Training Neural Networks: the Backpropagation Algorithm

Go to class Write review

Overview

This course dives into how neural networks learn from data. You'll implement loss functions to measure prediction errors, understand the intuition and mechanics of gradient descent, master the backpropagation algorithm to calculate gradients, and use an optimizer to update network weights.

Syllabus

Unit 1: Measuring Error: Loss Functions (MSE)

Mastering the MSE Loss Function
MSE Loss the Manual Way
Vectorized MSE for Neural Networks
Comparing MSE for Single and Batch

Unit 2: Gradient Descent: The Path to Minimizing Loss

Taking the First Step Downhill
Exploring the Power of Learning Rate
Keep the Gradient Moving
Stopping Gradient Descent at the Right Time
Writing the Heart of Gradient Descent

Unit 3: Backpropagation: Calculating Gradients in a Layer

Activation Derivatives in Backpropagation
Mastering Weight Gradients in Backpropagation
Building Gradients for Neural Layers
Gradient Shape Check in Backpropagation

Unit 4: Backpropagation Through the Multi-Layer Perceptron

Getting Gradients Going the Right Way
Backward Pass Through the Network
Build and Probe a Tiny MLP

Unit 5: Updating Weights with Stochastic Gradient Descent (SGD)

Making Neural Networks Learn
Mini Batch Magic for Neural Networks
Getting Gradient Steps Right
One Step of Neural Network Learning
Building the Neural Network Training Loop

Reviews

Start your review of Training Neural Networks: the Backpropagation Algorithm