Data Sanity Check for Deep Learning Systems via Learnt Assertions

SaneDL is a tool that provides systematic data Sanity check for Deep Learning-based Systems. SaneDL serves as a lightweight, flexible, and adaptive plugin for DL models. It is the first assertion-based tool that can automatically detects invalid input cases for DL systems. We show the effectiveness and efficiency of invalid input detection via real-world valid input cases, other than manipulated faulty samples typically used in the community. Our work can shed light to other practices in improving DL reliability.

Auto91

https://arxiv.org/abs/1909.03835