Many data science competitions suffer from a test set being markedly different from a training set (a violation of the “identically distributed” assumption). It is then difficult to make a representative validation set. We propose a method for selecting training examples most similar to test examples and using them as a validation set. The core of this idea is training a probabilistic classifier t
![Adversarial validation, part one - FastML](https://cdn-ak-scissors.b.st-hatena.com/image/square/a0bf28722eeec2a6bcbd1caa39738045c0077ad0/height=288;version=1;width=512/http%3A%2F%2Ffastml.com%2Fimages%2Fdata-projector%2Fsantander.png)