Deep Learning: ANN and CNN

Machine learning had three branches. Deep learning has two, and the syllabus names both. You are not asked to build either — only to recognise which one a description is talking about.

1Where they sit

Deep learning is the other half of the learning based branch, beside machine learning. It trains itself on huge amounts of data — such machines are intelligent enough to develop algorithms for themselves.

The deep learning branch

Switch between ANN and CNN. Both hang off Deep Learning, which hangs off Learning Based — not off Machine Learning.

Deep Learning. Modelled on the brain; extracts features by itself.

Full lineage: AI Models → Learning Based → Deep Learning → ANN

2Artificial Neural Network (ANN)

ANN

Artificial Neural Networks are modelled on the human brain and nervous system. They can extract features automatically, with no input from the programmer. Every node is essentially a machine learning algorithm, and the whole thing is useful when the dataset is very large.

3Convolutional Neural Network (CNN)

CNN

A CNN is a deep learning algorithm that takes an input image, assigns importance — learnable weights and biases — to the various aspects and objects in the image, and is thereby able to tell one from another.

4Telling them apart

	ANN	CNN
Modelled on	The human brain and nervous system	Not stated — it is built for pictures
Input it is built for	Any very large dataset	Images
The giveaway word	“inspired by the brain”, “each node is an ML algorithm”	“image”, “assigns importance to features in the picture”
Extracts features	Automatically, with no programmer input	Automatically, from the image

One word decides it

If the description mentions an image, it is a CNN. If it mentions the brain, it is an ANN. The board’s question in 2025–26 was built on exactly those two cues.

5From the board papers

Identify the type of deep learning model.
(a) This deep learning model processes input images by learning to assign importance (weights and biases) to various features in the image, allowing it to distinguish one object from another.
(b) This deep learning model is inspired by the structure of the human brain. It can automatically extract features from large datasets without explicit programming. Each node in the network acts like a small machine learning algorithm.CBSE sample paper 2025–26

2 marks

▶Show the answer

(a) Convolutional Neural Network (CNN). It takes images in and assigns weights and biases to what is in them.
(b) Artificial Neural Network (ANN). Inspired by the brain, extracts features by itself, and every node is a small machine learning algorithm.

6Check yourself

Quick Check

What are the two types of deep learning model named in the Class 10 syllabus?

Quick Check

Which deep learning model is built to take an image as its input?

Quick Check

In an ANN, what is every node essentially?

← Previous

🫧 Clustering and Association

What is a Neural Network? 🕸️