Neural Network Basics Quiz

What does one 'epoch' represent during the training of a neural network?

• One forward pass through a single data point

• One complete pass through the entire training dataset

• One update of the weights

• One evaluation on the test set

What are 'hidden layers' in a neural network?

• The input and output layers

• Layers between the input and output layers

• Layers used only for visualization

• Layers that store the final predictions

What is the purpose of an 'optimizer' in neural network training?

• To define the network architecture

• To adjust the weights to minimize the loss

• To evaluate the model performance

• To split the data into training and testing sets

What are 'weights' in a neural network?

• The input features

• The biases of the neurons

• Parameters that determine the strength of connections between neurons

• The activation functions applied to the neurons

What is 'bias' in a neuron?

• A measure of how often the neuron fires

• A constant value added to the weighted sum of inputs

• The learning rate of the neuron

• The number of connections to the neuron

What is the purpose of an activation function in a neuron?

• To calculate the loss

• To introduce non-linearity

• To adjust the weights

• To divide the input data

What is the role of a 'loss function' in training a neural network?

• To activate the neurons

• To measure the difference between predictions and actual values

• To optimize the weights

• To preprocess the input data

What does the term 'learning rate' refer to in the context of neural network training?

• The number of epochs

• The size of the dataset

• The step size at which the optimizer updates the weights

• The type of activation function used

What is the final layer in a typical classification neural network often called?

• Hidden layer

• Output layer

• Input layer

• Pooling layer

Which of the following is a technique used to prevent overfitting in neural networks?

• Increasing the learning rate

• Dropout

• Using a linear activation function

• Removing dropout