How to make it possible for a neural network to tune its own hyper parameters?

Question

I am curious about what would happen to hyperparameters when they would be set by a neural network itself or by creating a neural network that encapsulates and influences the hyperparameters of the network it encapsulates.

The goal for me here is to experiment and get some more in-depth knowledge about neural networks. But I had a hard time finding information to be able to execute such an experiment. Which point in a direction that either it has never been done before or the idea is just really dumb.

Now what I would like to know is, does anyone of you know where I would be able to find information (e.g. books, web-articles, papers, et cetera) to conduct such an experiment?