CS5720 - Week 1
Slide 6 of 20

The Perceptron - Simplest Neural Network

What is a Perceptron?

A Perceptron is the simplest form of a neural network, consisting of a single artificial neuron that makes binary decisions based on weighted inputs.
Key Characteristics:

Single Layer: One neuron, multiple inputs
Binary Output: Returns 0 or 1
Linear Classifier: Draws a line to separate data
Supervised Learning: Learns from examples
🎯 Think of it like...
A perceptron is like a bouncer at a club who makes yes/no decisions based on multiple criteria (age, dress code, ID) with different importance weights.

Perceptron Structure

x₁
x₂
1
w₁
w₂
b
Σ
f(∑)
y
Click on any component for detailed explanation!

The Perceptron Formula

y = f(w₁x₁ + w₂x₂ + b)
Inputs (x₁, x₂)
The data features or signals coming into the perceptron
Weights (w₁, w₂)
Importance multipliers for each input - learned during training
Bias (b)
Shifts the decision boundary - like an offset or threshold
Activation Function f()
Converts the weighted sum into a binary decision (0 or 1)
Prepared by Dr. Gorkem Kar