DSE DSE - Data Science Deep Learning and Neural Networks

Question 1

Which activation function is most commonly used in hidden layers of modern deep neural networks to mitigate the vanishing gradient problem?

Accepted Answer

ReLU

Answer

Sigmoid

Answer

Tanh

Answer

Softmax

Question 2

What regularization technique prevents overfitting in neural networks by randomly setting a fraction of neuron activations to zero during each training step?

Accepted Answer

Dropout

Answer

L2 regularization

Answer

Batch normalization

Answer

Early stopping

Question 3

Which optimization algorithm combines momentum with adaptive per-parameter learning rates and is widely used as a default optimizer for deep learning?

Accepted Answer

Adam

Answer

SGD

Answer

RMSProp

Answer

Adagrad

Question 4

What causes the 'vanishing gradient' problem in deep neural networks?

Accepted Answer

Gradients shrinking exponentially as they propagate backward through many layers with saturating activations

Answer

Too many neurons per layer

Answer

Using too large a learning rate

Answer

Overfitting on the training set

Question 5

Which neural network architecture was specifically designed and is primarily used for image recognition and classification tasks?

Accepted Answer

Convolutional Neural Network (CNN)

Answer

Recurrent Neural Network (RNN)

Answer

Generative Adversarial Network (GAN)

Answer

Transformer

Question 6

In a neural network, what is the purpose of the backpropagation algorithm?

Accepted Answer

Computing gradients of the loss with respect to each weight so they can be updated

Answer

Initializing weights before training

Answer

Normalizing input features

Answer

Selecting the optimal number of hidden layers