Deep Learning Quiz

Which company uses deep learning for facial recognition in photos?

• Amazon

• Facebook

• Google

• Netflix

What is the primary function of the Input Layer in a neural network?

• To process and classify data

• To receive input data and represent features of the data

• To adjust weights during training

• To generate the final output of the network

Which of the following is NOT a commonly used activation function in neural networks?

• Sigmoid

• ReLU (Rectified Linear Unit)

• Tanh

• Polynomial

What is the role of the learning rate in a neural network?

• It determines the number of neurons in the hidden layer.

• It controls how quickly or slowly the model learns.

• It decides the type of activation function used.

• It specifies the number of iterations for training.

What happens if the learning rate is set too high in a neural network?

• The model may skip the global minimum of the loss function.

• The model will learn very slowly.

• The number of neurons in the output layer will increase.

• The activation function will not work properly.

What is the formula for the Tanh activation function?

• tanh(x) = 2σ(2x) - 1

• σ(x) = 1 / (1 + exp(-x))

• ReLU(x) = max(0, x)

• Leaky ReLU(x) = max(0.01x, x)

What is the primary advantage of the ReLU activation function?

• Computational efficiency and no vanishing gradient problem

• Handles negative input values better

• Converts input into probability vectors

• Solves the "dying ReLU problem"

How does the Leaky ReLU activation function address the "dying ReLU problem"?

• By introducing small positive values for negative inputs

• By converting input into probability vectors

• By improving computational efficiency

• By eliminating the vanishing gradient problem

Which cost function is most commonly used for regression problems?

• Cross-Entropy

• Mean Squared Error (MSE)

• Binary Cross-Entropy

• Softmax Loss

What does the formula for Mean Squared Error (MSE) calculate?

• The average squared difference between predicted and actual values

• The sum of absolute differences between predicted and actual values

• The product of predicted and actual values

• The logarithmic difference between predicted and actual values