Perceptron Model

The Perceptron is one of the earliest and simplest types of artificial neural networks. It serves as the foundation for modern deep learning models. The perceptron is used for binary classification tasks, where the output is either 0 or 1. Understanding this model helps you grasp how neural networks learn and make decisions.

What is a Perceptron?
A perceptron is a single neuron model that takes multiple inputs, applies weights, adds a bias, and passes the result through an activation function to produce an output.

Mathematical Representation
The perceptron computes an output using the following formula:

• Weighted sum:
  z = (w1 × x1) + (w2 × x2) + … + (wn × xn) + b
• Activation function:
  output = f(z)

Where:

• x = input features
• w = weights
• b = bias
• f = activation function

Activation Function
The perceptron typically uses a step function as the activation function:

• Output = 1 if z ≥ 0
• Output = 0 if z < 0

This makes it suitable for binary classification problems.

Working of a Perceptron

1. Inputs are multiplied by their respective weights
2. The weighted values are summed along with a bias
3. The result is passed through the activation function
4. The final output is produced as 0 or 1

Training the Perceptron
The perceptron learns by adjusting its weights based on errors:

• Predict the output for given inputs
• Compare with actual output
• Update weights using a learning rule
• Repeat until the model performs well

Perceptron Learning Rule
Weights are updated using the formula:
w = w + learning_rate × (actual − predicted) × input

This process helps the model improve its predictions over time.

Example in Python

import numpy as np# Sample data
X = np.array([[0,0], [0,1], [1,0], [1,1]])
y = np.array([0, 0, 0, 1])# Initialize weights and bias
weights = np.zeros(2)
bias = 0
learning_rate = 0.1# Training loop
for epoch in range(10):
    for i in range(len(X)):
        z = np.dot(X[i], weights) + bias
        prediction = 1 if z >= 0 else 0
        error = y[i] - prediction
        weights += learning_rate * error * X[i]
        bias += learning_rate * errorprint("Weights:", weights)
print("Bias:", bias)

Limitations of Perceptron

• Can only solve linearly separable problems
• Cannot handle complex patterns like XOR
• Limited to binary classification tasks

Importance in Deep Learning

• Forms the building block of neural networks
• Helps understand weights, bias, and activation functions
• Leads to advanced models like multi-layer perceptrons

Lesson Summary
In this lesson, you learned about the perceptron model, its structure, mathematical formulation, training process, and limitations. The perceptron is a fundamental concept that lays the groundwork for understanding more complex neural networks in deep learning.