NCA AI & Deep Learning Frameworks

Question 1

What is the purpose of a neural network in deep learning?

Accepted Answer

To model complex data relationships

Answer

To store data.

Answer

To manage hardware.

Answer

To encrypt data.

Question 2

Which activation function outputs values between 0 and 1?

Accepted Answer

Sigmoid

Answer

ReLU.

Answer

Tanh.

Answer

Linear.

Question 3

What is overfitting in deep learning?

Accepted Answer

Model learns noise and poorly generalizes

Answer

Model generalizes well.

Answer

Model performs poorly on training data.

Answer

Model is undertrained.

Question 4

Which framework is developed by Google for machine learning?

Accepted Answer

TensorFlow

Answer

PyTorch.

Answer

Caffe.

Answer

Theano.

Question 5

What is backpropagation used for?

Accepted Answer

Weight updates using gradients

Answer

Data preprocessing.

Answer

Model deployment.

Answer

Data visualization.

Question 6

Which layer type is commonly used in Convolutional Neural Networks (CNNs)?

Accepted Answer

Convolutional layer

Answer

Fully connected layer.

Answer

Recurrent layer.

Answer

Dropout layer.

Question 7

What is a common technique to prevent overfitting?

Accepted Answer

Dropout

Answer

Increasing learning rate.

Answer

Decreasing batch size.

Answer

Using only training data.

Question 8

Which library is popular for deep learning with Python?

Accepted Answer

PyTorch

Answer

Scikit-learn.

Answer

NumPy.

Answer

Pandas.

Question 9

What does the term 'epoch' mean in deep learning training?

Accepted Answer

One full training data pass

Answer

Single batch processed.

Answer

Model initialization.

Answer

Validation phase.