How to Estimate the Performance of a Machine Learning Model When You Don't Have Ground Truth

There are a number of different algorithms that you can use when you don't have ground truth in order to estimate the performance of a machine learning model. The one that I think has the most academic rigor behind it and looks the most interesting is one called mandolin. So what mandolin does is it basically slices and dices your data in a bunch of different ways. And then allows you to create estimates of if I see my data drifting in this feature by this much, this probably has this kind of impact on myMachine Learning model. You know, we need to use something like mandolin to be able to estimate that, but at least we know that if there are changes,. those