Artificial Intelligence Neural Networks

Question 1

What is the role of an activation function in a neural network?

Accepted Answer

Introduce non-linearity into the model

Answer

Initialize the weights

Answer

Calculate the loss

Answer

Normalize the input data

Question 2

Which activation function outputs values between 0 and 1 and is commonly used in binary classification output layers?

Accepted Answer

Sigmoid

Answer

ReLU

Answer

Tanh

Answer

Softmax

Question 3

What is backpropagation in neural network training?

Accepted Answer

Using the chain rule to compute gradients and update weights

Answer

Feeding data forward through the network

Answer

Randomly initializing network weights

Answer

Normalizing layer activations

Question 4

What is a 'vanishing gradient' problem in deep neural networks?

Accepted Answer

Gradients shrink to near zero as they propagate back, slowing learning in early layers

Answer

Gradients become extremely large, destabilizing training

Answer

The network loses memory of earlier inputs

Answer

Weights are initialized too close to zero

Question 5

Which layer type in a neural network is fully connected, meaning every input neuron links to every output neuron?

Accepted Answer

Dense (fully connected) layer

Answer

Convolutional layer

Answer

Pooling layer

Answer

Embedding layer

Question 6

What does 'dropout' do during neural network training?

Accepted Answer

Randomly deactivates neurons during each training step to reduce overfitting

Answer

Removes neurons permanently to reduce model size

Answer

Scales learning rate based on layer depth

Answer

Normalizes output of each layer