← All lessons
Week 1 · Foundations of Generative AI
MAM5020F — Generative AI for Research

But What is a Neural Network?

A Visual Introduction to Deep Learning
Contents
← All lessons
Week 1 · Foundations of Generative AI
Key terms
Neural network
A computational system loosely inspired by the brain, made of layers of simple mathematical units (neurons) connected together. By adjusting connection strengths through training, it learns to recognise patterns in data.
Deep learning
A branch of machine learning using neural networks with many layers, whose depth lets the system learn increasingly abstract features from raw data, powering most modern AI including language and image models.
Transformer
A neural network architecture introduced in 2017 that processes whole sequences of text at once and uses attention to weigh how relevant each word is to every other. It underpins virtually all modern large language models.

Why Watch This Video?

This is one of the clearest introductions to neural networks available online. Grant Sanderson (3Blue1Brown) makes complex mathematical concepts visually intuitive and accessible to everyone, from high school students to professional researchers.

Instead of opening with intimidating equations, this video builds understanding from the ground up using animations and clear analogies. You'll see how neural networks learn to recognize patterns, understand why they're structured the way they are, and get a feel for what "deep learning" actually means.

No advanced mathematics required. If you can follow basic arithmetic and are willing to think visually, you'll understand neural networks by the end of this 19-minute video.

After Watching

After watching this video, you'll have the conceptual foundation to understand how modern AI systems work. The transformers, diffusion models, and GANs we discussed earlier all build on these same basic principles. They are more sophisticated arrangements of the same fundamental building blocks.

Optional: If you found this video helpful and want to go deeper, the remaining videos in the series (gradient descent, backpropagation, and what neural networks actually learn) provide even richer understanding. But this first video alone gives you 80% of what you need to be an informed AI user in research.

← PreviousA Lightning Tour of AI (Video)

Drafted with Claude (Anthropic) and reviewed by Jonathan Shock before publication. AI-generated errors are possible — if you spot one, please email jonathan.shock@uct.ac.za. Full detail: AI Content Disclaimer.

© 2026 Jonathan Shock · MAM5020F: Generative AI for Research · CC BY 4.0

Week 1
Lesson 4
1 min read
Key terms
Neural network
A computational system loosely inspired by the brain, made of layers of simple mathematical units (neurons) connected together. By adjusting connection strengths through training, it learns to recognise patterns in data.
Deep learning
A branch of machine learning using neural networks with many layers, whose depth lets the system learn increasingly abstract features from raw data, powering most modern AI including language and image models.
Transformer
A neural network architecture introduced in 2017 that processes whole sequences of text at once and uses attention to weigh how relevant each word is to every other. It underpins virtually all modern large language models.