1. What is forward propagation in a neural network?

It involves passing inputs through the network to compute the output. It updates the weights based on the gradient of the loss function.

2. Which of the following describes backward propagation?

Calculating the loss of the model. Updating weights by propagating the gradient backward through the network. Computing the output for given inputs.

3. What does the activation function do in a neural network?

Introduces non-linearity to the model. Optimizes the weights during training. Connects input nodes to output nodes directly.

4. Which of the following statements is true about gradient descent?

It is used to minimize the loss function by adjusting weights. It increases the loss function to improve learning. It is unrelated to backpropagation.

5. What does the term 'learning rate' refer to in a neural network?

The step size used during gradient updates. The total number of layers in the network. The number of training iterations.

6. What is the purpose of the loss function in training a neural network?

To evaluate the accuracy of the model. To quantify how far the model's predictions are from the actual values. To initialize the weights of the model.

7. In backpropagation, what does the gradient represent?

The amount by which the weights should be adjusted. The output of the activation function. The final prediction of the network.

8. What is the chain rule used for in backpropagation?

To compute the total output of the network. To calculate the gradients of composite functions with respect to their inputs. To initialize the neural network's weights.

9. Which component is adjusted during training to reduce the loss function?

The input data. The model's weights. The activation function.

10. What is the final output of a neural network called in a classification task?

The gradient. The loss value. The predicted class or probability.