Artificial Intelligence Neural Networks and Deep Learning

Question 1

What is the activation function that outputs values strictly between 0 and 1, commonly used in binary classification output layers?

Accepted Answer

Sigmoid

Answer

ReLU

Answer

Tanh

Answer

Softmax

Question 2

Which layer type in a convolutional neural network applies a learned filter to detect local patterns in an image?

Accepted Answer

Convolutional layer

Answer

Fully connected layer

Answer

Pooling layer

Answer

Dropout layer

Question 3

What problem does the 'vanishing gradient' phenomenon cause in deep neural networks?

Accepted Answer

Gradients become very small in early layers, halting learning

Answer

The model converges too quickly

Answer

The network generates random outputs

Answer

Training data is too small to support the model

Question 4

What is the purpose of a dropout layer in a neural network?

Accepted Answer

To randomly deactivate neurons during training to reduce overfitting

Answer

To speed up forward passes

Answer

To normalize activations across a batch

Answer

To increase the number of parameters

Question 5

Which optimizer adapts the learning rate for each parameter based on historical gradient information?

Accepted Answer

Adam

Answer

SGD

Answer

Momentum

Answer

Batch gradient descent

Question 6

What does 'backpropagation' compute in a neural network training cycle?

Accepted Answer

The gradient of the loss with respect to each network weight

Answer

The forward pass predictions

Answer

The optimal batch size for training

Answer

The number of layers needed