Thread by @svpino

I've heard multiple times that you don't need to do any feature engineering or selection whenever you are using neural networks.

This is not true.

Yes, neural networks can extract patterns and ignore unnecessary features from the dataset, but this is usually not enough.

↓

First, neural networks can't compete with our expertise understanding the data.

As long as we know the problem and the dataset, we can come up with features that it would be really hard for a network to reproduce.

This is a highly creating process. Really hard to reproduce.

A couple of notes regarding the ability of a network to do automatic feature selection:

Yes, networks can "ignore" features that have no bearing on the prediction.

But these features can still introduce noise that degrade the performance of the model.

Unnecessary features add complexity to the entire process.

You need to collect, maintain, and store them.

On top of that, unnecesary features increase the required amount of data, training time, memory requirements, and inference time.

They are dead weight.

Feature engineering and feature selection are key steps in the machine learning process regardless of which model you decide to use.

Neural networks help, but they won't do the job for you.