Vladyslav Kolbasin
14 Oct 2018
10:55 - 11:40
Stream B

How to explain predictions of your network?

In this talk we are going to take a look at interpretable models, why we need them and what it actually means to interpret a model. I will give an overview of the approaches and libraries for models’ explanation and analysis both in the white-box and in the black-box representation. We are also going to take a look at the example of how analysis helps to improve the quality of a specific model.