KEMBAR78
Artificial Intelligence Lab Assignment1 | PDF | Algorithms | Machine Learning
Scribd
Upload
12 views3 pages

Artificial Intelligence Lab Assignment1

The document outlines an assignment on the design and implementation of Fully Connected Neural Networks (FCNN) by Md. Mahfuzar Rahman from the University of Rajshahi. It details the architecture of the FCNN, including the roles of input, hidden, and output layers, as well as the number of neurons and parameters in each layer. Additionally, it provides resources for further exploration, including a GitHub repository and a Google Colab notebook for FCNN implementation.

Uploaded by

Md Mahfuzar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
12 views3 pages

Artificial Intelligence Lab Assignment1

The document outlines an assignment on the design and implementation of Fully Connected Neural Networks (FCNN) by Md. Mahfuzar Rahman from the University of Rajshahi. It details the architecture of the FCNN, including the roles of input, hidden, and output layers, as well as the number of neurons and parameters in each layer. Additionally, it provides resources for further exploration, including a GitHub repository and a Google Colab notebook for FCNN implementation.

Uploaded by

Md Mahfuzar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Artificial Intelligence Lab

Assignment 1
Design and Implementation of Fully Connected Neural
Networks

Md. Mahfuzar Rahman


ID: 2110976123
Session: 2020–21
Department: Computer Science and Engineering
University of Rajshahi

Fully Connected Neural Network Diagram

Figure 1: Fully Connected Neural Network Architecture


Explanation of the Diagram
The figure above represents a Fully Connected Neural Network (FCNN),
where each neuron in a layer is connected to all neurons in the next layer
through weighted connections.
Layer Roles:

• Input Layer: - Accepts the input features from the dataset. - In this
network, there are two input neurons (x1 and x2 ). - The input layer
does not perform computation; it simply forwards the feature values to
the first hidden layer.

• Hidden Layers: - Perform computations to transform inputs into


more useful intermediate representations. - Each neuron calculates a
weighted sum of inputs plus a bias term:
n
X
z= w i · xi + b
i=1

- An activation function f (z) is then applied to introduce non-linearity,


allowing the network to learn complex patterns. - In this architecture:

– Hidden Layer 1: 2 neurons (f1 |g1 and f2 |g2 ).


– Hidden Layer 2: 3 neurons (f3 |g3 , f4 |g4 , f5 |g5 ).
– Hidden Layer 3: 2 neurons (f6 |g6 and f7 |g7 ).

• Output Layer: - Produces the final output prediction from the net-
work. - Here, it has one neuron (f8 |g8 ) that combines the outputs of
the last hidden layer. - Depending on the activation function (e.g.,
sigmoid, softmax, linear), it can output probabilities (classification) or
continuous values (regression).

2
Model Summary Output

Figure 2: Model Summary of the Fully Connected Neural Network

Explanation of Model Summary


The model consists of:

• Input Layer: Shape (N one, 2), 0 parameters.

• Dense Layer 1: 2 neurons, 6 parameters (4 weights + 2 biases).

• Dense Layer 2: 3 neurons, 9 parameters (6 weights + 3 biases).

• Dense Layer 3: 2 neurons, 8 parameters (6 weights + 2 biases).

• Output Layer: 1 neuron, 3 parameters (2 weights + 1 bias).

• Total Parameters: 26 trainable parameters.

Resources
• GitHub Repository: GitHub - Artificial Intelligence Lab

• Google Colab Notebook: Colab - FCNN Implementation

You might also like