For linear regression, one can calculate a confidence interval for a prediction. Is this possible for all machine learning techniques?
For example let's say I fit a neural network, and then start using it to make predictions on new data. Is it possible to create a confidence interval around my prediction? Perhaps using the distribution of the cross validation errors? Or perhaps using an entirely new machine learning model to predict the error?
I'm really try to wrap my head around all of this. I really appreciate any insight.