---
title: "Role of perceptron in neural network"
subject: "Artificial Intelligence"
chapter: 6
key: "semester_6/artificial_intelligence:perceptron-role"
source: https://www.bctengineer.com/q/semester_6/artificial_intelligence/perceptron-role
site: BCT Engineering
---

# Role of perceptron in neural network

Artificial Intelligence · Chapter 6: Machine Learning

- Asked in: 2069 Poush, 4 marks; 2070 Magh, 3 marks (asked 2 times)
- Video: [Perceptron | Neural Networks](https://www.youtube.com/watch?v=OFbnpY_k7js) — First Principles of Computer Vision
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_6/artificial_intelligence/perceptron-role
- Whole chapter: https://www.bctengineer.com/md/semester_6/artificial_intelligence/chapter-6

## Question

What is the role of perceptron in neural network?

## Short answer

The perceptron is the simplest neural network, a single neuron with a step activation that acts as a binary linear classifier. It serves as the basic building block for multilayer networks, implements logic gates like AND/OR, and its learning rule was foundational. Its inability to solve XOR motivated the development of multilayer networks and back-propagation.

## Answer

A **perceptron** (Rosenblatt, 1958) is the simplest neural network: a single artificial neuron that computes a weighted sum of its inputs and applies a threshold (step) activation to give a binary output.

y = 1 if Σ wᵢxᵢ + b > 0, otherwise 0 (or −1).

```text
 x1 --w1-->\
 x2 --w2--->( Σ + b )--> step --> y
 xn --wn-->/
```

#### Role of the perceptron in neural networks
1. **Basic building block:** every unit of a multilayer network is a generalisation of the perceptron (with a smooth activation); networks of perceptrons form multilayer perceptrons (MLP).
2. **Binary classifier:** it separates two classes with a linear decision boundary (a line or hyperplane), so it solves linearly separable problems such as AND and OR.
3. **Learning model:** the perceptron learning rule w ← w + η(t − y)x was one of the first supervised learning algorithms. It converges if the classes are linearly separable.
4. **Logic gates:** a single perceptron can implement AND, OR and NOT.
5. **Historic role:** its limitation (it cannot learn XOR) led to multilayer networks and back-propagation.

**Limitation:** a single perceptron cannot solve non-linearly separable problems; hidden layers are needed.

## Check yourself

The answer's key points as multiple-choice questions, with the correct option and why.

1. How does a perceptron decide its output?
   A. It outputs 1 if the weighted sum plus bias is greater than 0, otherwise 0 (or -1)
   B. It outputs the weighted sum without any threshold
   C. It outputs 1 only if all inputs are 1
   D. It outputs a value between 0 and 1 using a sigmoid
   Answer: A. A perceptron uses a step (threshold) activation giving a binary output. (Key point: Perceptron output)

2. What kind of problems can a single perceptron solve?
   A. Any problem including XOR
   B. Linearly separable problems such as AND and OR
   C. Only problems with three or more classes
   D. Only non-linear problems
   Answer: B. It forms a linear boundary (line or hyperplane), so XOR is beyond it. (Key point: Decision boundary)

3. What is the perceptron learning rule?
   A. w <- w + eta * y
   B. w <- w - eta * x * x
   C. w <- w + eta(t - y)x
   D. w <- w * (t - y)
   Answer: C. Weights change in proportion to the error (t - y) and the input x; it converges for linearly separable data. (Key point: Learning rule)

4. Why was the perceptron's limitation historically important?
   A. It proved neural networks are useless
   B. It led to the invention of the step function
   C. It showed XOR can be solved by a single neuron
   D. Its inability to learn XOR led to multilayer networks and back-propagation
   Answer: D. Hidden layers were introduced to handle non-linearly separable problems. (Key point: Historic role)

## Related in Artificial Intelligence

- [Perceptron; neural network as OR gate](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Aperceptron-or-gate) — Artificial Intelligence, chapter 6, asked 2×
- [Perceptron; neural network as XOR gate](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Aperceptron-xor-gate) — Artificial Intelligence, chapter 6, asked 2×
- [Neural network, back-propagation, perceptron](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Ann-bp-perceptron) — Artificial Intelligence, chapter 6
- [Neural network; supervised vs unsupervised](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Ann-supervised-unsupervised) — Artificial Intelligence, chapter 6
- [Neural network; back-propagation](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Aneural-network-bp-algorithm) — Artificial Intelligence, chapter 6

## Related in other subjects

- [Short note: Neural network, perceptron rule](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Aneural-network-perceptron-note) — Data Mining, chapter 3