ANN and Application

What is the name of this function ?

• 1. Mean Squared Error

• 1. Cross Entropy 

• Softmax

• Softmax

Which statement is correct?

• 1. If the number of neurons in a hidden layer is too small, it may cause overfitting

• The number of neurons in the output layer is equal to the number of classes

• A low weight value will significantly change the output.

• All statements are correct

How many machine learning types are there?

• 6

• 3

• 4

• 5

How far in advance can ANN technology predict the weather?

• a week

• a month

• two weeks

• 50 days

Recurrent Neural Networks (RNN) are usually used for facial recognition and image processing

• True

• False

Which fields that ANN have contributed to the aerospace development?

• Auto Piloting

• Modeling key dynamic simulations

• Fault diagnosis

• All of above

Why is backpropagation needed in a neural network?

• To achieve data input for NN
  

• To minimize loss function in NN

• To minimize loss function in NN and achieve data for input

• To minimize loss function in NN and improve prediction accuracy

What are the types of backpropagation?

• Static and Through time

• Recursive

• Static and Non-static

• Feedforward and Recursive

 Which one is one of the disadvantages of backpropagation?

• Adjust too many parameters after each repeat

• Time-consuming

• Altering output accuracy from changing weights

• Not flexible

 How can a NN reduce error with backpropagation?

• Remember the output

• Find better fitting input

•  Adjust its’ weight

• Update its’ database