Data Science and Machine Learning (Theory and Projects) A to Z - DNN and Deep Learning Basics: DNN Architecture Exercise

The video tutorial explains how to calculate the total number of weights or parameters in a deep neural network. It covers the step-by-step process of determining weights for each layer, including the input, hidden, and output layers. The tutorial also discusses the implications of having a large number of weights, such as increased model complexity and the risk of overfitting, especially with limited training data. The importance of considering the number of parameters when designing a neural network architecture is emphasized.

How many weights does a single neuron have in a fully connected setup if it receives four inputs?

• 4

• 2

• 5

• 3

What is the total number of weights for the first layer if there are 5 neurons, each receiving 4 inputs?

• 16

• 28

• 20

• 24

In the second layer, how many weights does each neuron have if it receives inputs from 5 neurons in the previous layer?

• 5

• 6

• 7

• 4

How many weights are there in total for the output layer if it consists of 2 neurons, each receiving 5 inputs?

• 12

• 10

• 14

• 8

What is a potential risk of having too many weights in a neural network model?

• Underfitting

• Simpler model

• Faster training

• Overfitting