Scribd
Upload
8 views7 pages

Module 1 Solved

The document outlines calculations for the outputs of hidden layer neurons using sigmoid activation functions based on given inputs, weights, and biases. It includes multiple scenarios with different input vectors and assumptions about weights, leading to final output values. Additionally, it mentions performing forward and backward passes in a neural network context, with a focus on weight updates based on errors.

Uploaded by

blessytsunny
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
8 views7 pages

Module 1 Solved

The document outlines calculations for the outputs of hidden layer neurons using sigmoid activation functions based on given inputs, weights, and biases. It includes multiple scenarios with different input vectors and assumptions about weights, leading to final output values. Additionally, it mentions performing forward and backward passes in a neural network context, with a focus on weight updates based on errors.

Uploaded by

blessytsunny
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

Module 1

Part A

1.
Suppose you have a 3-dimensional input x = (x1, x2, x3) = (2, 2, 1) fully
connected to 1 neuron which is in the hidden layer with activation function
sigmoid. Calculate the output of the hidden layer neuron. Let W = (0.5, -1, 2)
and bias = -1. (Aug 2024)

1. Output of the Hidden Layer


Neuron
We compute this in two steps:

Step 1: Weighted sum (z)

Step 2: Apply Sigmoid Activation

Sigmoid function:

Final Output = 0.5

A 3-dimensional input X = (X1, X2, X3) = (1, 2, 1) is fully connected to 1


2. neuron which is in the hidden layer with binary sigmoid activation function.
Calculate the output of the hidden layer neuron. Assume associated
weights. Neglect bias term. (Oct 2023)
If weights are not given, clearly state your assumption.

Assume weights:
W=(1,1,1),bias = 0 (neglected)

Given:
X=(1,2,1)

Step 1: Compute weighted sum

Step 2: Apply Sigmoid

Final Output ≈ 0.982

3. Calculate the output Y of a three input neuron with bias. The input feature
vector is (x1, x2, x3) = (0.8, 0.6, 0.2) and weight values are [w1, w2, w3] = [0.2,
0.1, -0.3] and Bias is 0.25. Use Binary Sigmoid as the activation function.
(May 2024)

Step 1: Compute weighted sum (z)

Substitute values:

Step 2: Apply Binary Sigmoid Activation

Final Output
Y ≈ 0.60
4. A 3-dimensional input X = (X1, X2, X3) = (2, 1, 2) is fully connected to 1 neuron
which is in the hidden layer with binary sigmoid activation function. Calculate
the output of the hidden layer neuron. Assume associated weights. Neglect
bias term.

Since weights are not given, the most standard assumption is:

W=(1,1,1), bias = 0 (neglected)

Step 1: Compute weighted sum

Step 2: Apply Sigmoid

Final Output

Y ≈ 0.993

Part B

5.
Assume that neurons have Sigmoid activation function. Perform a forward
pass and backward pass on the network. Assume the actual output at node
6 is 1 and learning rate is 0.9.(August 2024) (14 marks)
Given
Inputs:
Activation: Sigmoid

1. Forward Pass
Hidden Neuron 4:

Hidden Neuron 5:

Output Neuron 6:

2. Backward Pass

Output Layer Error:


3. Weight Updates
Final Summary :

You might also like