Neural Networks
This page assumes nothing, not that you know what a neuron is, not that you remember any math from school. Every idea gets explained in plain language and an animation first, then in numbers, then you try it yourself. Everything below is live - touch it.
Foundations
3- 01
What Is a Neural Network?
A first, no-math mental model for what a neural network is and why it can learn from examples.
- 02
The Single Neuron
The building block: how one neuron computes a weighted sum, a bias, and an activation.
- 03
Layers and the Forward Pass
Why one neuron isn't enough, how stacking neurons into layers builds real capacity, and what a forward pass is.
Activation Functions
2Training a Network
7- 01
Loss Functions
How a network measures how wrong a prediction is, covering squared error and cross-entropy loss.
- 02
Gradients and Gradient Descent
What a gradient is and how repeatedly stepping downhill trains every weight in a network.
- 03
Optimizers
How optimizers like momentum, RMSProp, and Adam improve on plain gradient descent.
- 04
Weight Initialization
Why weights can't all start identical, and what breaks when naive random initialization is used at scale.
- 05
Xavier and He Initialization
Xavier and He initialization, the fixes that keep signal variance stable as a network gets deeper.
- 06
Regularization
Why models overfit, the bias-variance tradeoff, and how L1/L2 penalties pull a network back toward simplicity.
- 07
Training-Time Regularization
Training-procedure techniques - dropout, early stopping, batch normalization, and data augmentation - that fight overfitting without changing the loss formula.