A neural network is different from a multiple regression model because:

I. It has more layers such as a hidden layer.
II. The temptation to overfit can be weaker in a regression model.
III. It can deal with nonlinearities.

Correct Answer: I and III

A neural network takes an input, passes it through multiple layers of hidden neurons (mini-functions with unique coefficients that must be learned), and outputs a prediction representing the combined input of all the neurons. Ut can in principle model nonlinearities automatically.

The caveat: the temptation to overfit can be (even) stronger in neural networks than in regression, since adding hidden layers or neurons looks harmless. So be extra careful to look at out-of-sample prediction performance.