Perceptron Learning Algorithm

A Perceptron is the simplest neural network model. It is a binary linear classifier that learns a decision boundary by updating weights based on classification errors.

Definition
A perceptron computes a weighted sum of inputs and applies a sign function.
Learning Rule
Weights are updated only when prediction is incorrect.
Limitation
Can only solve linearly separable problems.
Why It Matters
The perceptron is the foundation of modern deep learning.

Controls

Legend
🟢 Green → Class +1 (True)
🔴 Red → Class -1 (True)
🔵 Blue → Predicted +1
🟠 Orange → Predicted -1

Click the button to train a new perceptron with random initial weights.