p$.$ Select which are the good practices to avoid overfitting include:

i$.$ Using a two part cost function which includes a regularizer to penalize

model complexity.

ii$.$ Using a good optimizer to minimize error on training data.

iii. Building a structure of nested subsets of models and train learning

machines in each subset, starting from the inner subset, and stopping

when the cross$-$validation error starts increasing.

iv$.$ Discarding $50\%$ of randomly chosen samples