Skip to main content
AIAIByDM/ai
LearnSkillsToolsGamesExamsNewsletterCommunity
Start learning
LearnSkillsToolsGamesExamsNewsletterCommunitySearch
Start learning
AI From Scratch/Phase 03/Lesson 03/~120 minutes

Backpropagation from Scratch

BuildPythonNo prerequisites

Backpropagation is the algorithm that makes learning possible. Without it, neural networks are just expensive random number generators.

Back to phaseDeep Learning CoreNext lessonActivation Functions

Phase 03

Deep Learning Core

01The Perceptron02Multi-Layer Networks and Forward Pass03Backpropagation from Scratch04Activation Functions05Loss Functions06Optimizers07Regularization08Weight Initialization and Training Stability09Learning Rate Schedules and Warmup10Build Your Own Mini Framework11Introduction to PyTorch12Introduction to JAX13Debugging Neural Networks
Loading lesson page...

Progress

0 / 13 phase lessons0%
Continue target

Phase 03

Deep Learning Core

01The Perceptron02Multi-Layer Networks and Forward Pass03Backpropagation from Scratch04Activation Functions05Loss Functions06Optimizers07Regularization08Weight Initialization and Training Stability09Learning Rate Schedules and Warmup10Build Your Own Mini Framework11Introduction to PyTorch12Introduction to JAX13Debugging Neural Networks

On this page

Resources

Gradient DebuggerPromptmain.jlCodemain.pyCodeSource lessonSource