Quiz

Insufficient model complexity.

Too much data.

Excessive model complexity.

High dropout rate.

L1 regularization (Lasso).

L2 regularization (Ridge).

Batch normalization.

Increasing the number of layers in the network.

The training loss increases rapidly.

Both training and validation losses continue to decrease indefinitely.

Training loss tends to plateau or decrease slower, mirroring the validation loss.

Validation loss increases significantly.

It decreases the complexity of the model.

It applies penalties to the loss function.

It diminishes overfitting by normalizing layer inputs.

It increases the number of neurons in the network.

Before the dense layer.

After the dense layer but before the activation function.

After the activation function.

At the end of the network.

Increases all weights equally.

Reduces some weights to zero.

Scales up smaller weights.

Keeps all weights constant.

To control the strength of the regularization.

To determine the learning rate.

To set the number of layers in the model.

To choose between L1 and L2 regularization.