How Do Neural Networks Work? #underthehood
summary
Explains how feedforward multilayer neural networks make predictions and how they are trained, using both mathematical formulas and code. Covers what it means for a network to "learn" and the role of tensors in the computations, using a standard multilayer network as the running example.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 12:24What are neural networks?
- 25:30Linear separation and first models
- 31:53Error function and learning goal
- 38:57Gradient descent explained
- 48:06Backpropagation in detail
- 55:35Summary
- 1:05:45Live demo and code
