Parts of a Neural Network
Class 12 · Artificial Intelligence
6.2 Parts of a Neural Network
A Neural Network is organised into layers of interconnected nodes called artificial neurons. Each layer performs a specific role in processing information.
Every basic neural network consists of three essential parts: the Input Layer, Hidden Layer(s) and Output Layer.
The basic flow of information in a neural network is:
Input Layer → Hidden Layer(s) → Output Layer
1. Input Layer
The Input Layer is the first layer of a neural network. It consists of units that represent the input fields or features of the dataset.
Each input unit corresponds to a specific feature or attribute of the problem being solved.
Suppose a neural network is being developed to analyse medical information. Features such as age and blood pressure can be provided as inputs to the network.
2. Hidden Layers
Hidden Layers are located between the input layer and the output layer. They contain nodes that process the input data.
A neural network may contain one or more hidden layers. The nodes within these layers are interconnected, and each connection is assigned a weight.
A weight represents the strength or importance of a particular connection between neurons.
| Feature | Description |
|---|---|
| Location | Between the input layer and output layer. |
| Function | Processes input information and helps identify patterns. |
| Number | A neural network may have one or more hidden layers. |
| Connections | Nodes are interconnected through weighted connections. |
3. Output Layer
The Output Layer is the final layer of a neural network. It consists of one or more units representing the target field(s).
The output units generate the final prediction or result produced by the neural network.
If a neural network is trained to predict whether a student requires additional academic support, the output layer may provide the final prediction based on the information processed by the network.
Parts of a Neural Network at a Glance
| Part | Position | Main Function |
|---|---|---|
| Input Layer | First layer | Receives the input features or attributes. |
| Hidden Layer(s) | Between input and output layers | Processes input data and identifies patterns. |
| Output Layer | Final layer | Produces the final prediction or result. |
Deep Neural Network
A neural network that contains two or more hidden layers is known as a Deep Neural Network (DNN).
The term deep refers to the depth of the network, particularly the number of hidden layers used for processing information.
More Hidden Layers → Deeper Network
A Deep Neural Network contains two or more hidden layers.
Competency-Based Question
A school develops a neural network to predict whether a student is likely to require additional academic support. The network receives attendance, assessment performance and other relevant features as inputs.
Identify the role of the input layer, hidden layers and output layer in this system.
Click to View Answer
The input layer receives the student's relevant features such as attendance and assessment performance. The hidden layers process these inputs and identify patterns in the data. The output layer produces the final prediction regarding the student's possible need for additional academic support.
Common Beginner Mistakes
- Confusing the input layer with the hidden layer.
- Thinking that the hidden layer produces the final result.
- Forgetting that the output layer is the final layer.
- Assuming that every neural network has only one hidden layer.
- Confusing a neuron or node with an entire layer.
- Forgetting that connections between nodes have associated weights.
- Confusing a Deep Neural Network with a network containing only one hidden layer.
Quick Revision
- A neural network consists of interconnected artificial neurons arranged in layers.
- The three essential parts are the Input Layer, Hidden Layer(s) and Output Layer.
- The Input Layer receives the features or attributes of the dataset.
- The Hidden Layers process the input information and help identify patterns.
- The Output Layer produces the final prediction or result.
- Connections between neurons have associated weights.
- A neural network with two or more hidden layers is called a Deep Neural Network (DNN).
Memory Trick
I → H → O
Remember:
- I → Input Layer
- H → Hidden Layer(s)
- O → Output Layer
Input → Process → Output
Exam Tips
- Learn the three basic parts of a neural network in the correct order: Input → Hidden → Output.
- Remember that the input layer represents the input features or attributes.
- Mention that hidden layers process information and identify patterns.
- Remember that the output layer represents the target field(s) and produces the final prediction or result.
- For questions on weights, write that they represent the strength or importance of connections between neurons.
- For questions on Deep Neural Networks, remember: two or more hidden layers.
- Draw the sequence Input → Hidden → Output whenever a diagram-based question is asked.
Frequently Asked Questions (FAQs)
1. What are the three main parts of a neural network?
The three main parts are the Input Layer, Hidden Layer(s) and Output Layer.
2. What is the function of the input layer?
The input layer consists of units representing the input fields or features of the dataset and receives the input information.
3. What is the function of hidden layers?
Hidden layers process the input data and help the neural network identify patterns. A neural network may contain one or more hidden layers.
4. What is the function of the output layer?
The output layer consists of one or more units representing the target fields and generates the final prediction or result.
5. What do weights represent in a neural network?
Weights represent the strength or importance of the connections between neurons.
6. What is a Deep Neural Network?
A Deep Neural Network (DNN) is a neural network containing two or more hidden layers.
7. What does the term "deep" refer to in Deep Learning?
The term deep refers to the depth of a neural network, particularly the number of hidden layers.
Summary
- A neural network consists of interconnected artificial neurons arranged in layers.
- The three essential parts are the Input Layer, Hidden Layer(s) and Output Layer.
- The input layer represents the input fields or features of the dataset.
- Hidden layers process the input data and help identify patterns.
- The output layer represents the target field(s) and produces the final prediction or result.
- Connections between neurons are assigned weights representing their strength or importance.
- A neural network containing two or more hidden layers is called a Deep Neural Network (DNN).
Next Topic: Components of a Neural Network