If you are interested in machine learning safety research topics such as out-of-distribution detection, anomaly detection, novelty detection, and open-set recognition and have been confused about how these topics are relevant to each other. Also if you are curious to know more about the ongoing challenges and open problems in this research area, Please read our latest accepted work on Transaction on Machine Learning (TMLR) which provides a comprehensive survey on these topics under the umbrella of a unified perspective.

Paper: https://lnkd.in/esjMZhri
Source code: https://lnkd.in/epbVxiCY

This is a joint research work with Mohammadreza Salehi, Hossein Mirzaei, Dan Hendrycks, Sharon Yixuan Li, Mohammad Hossein Rohban, and Mohammad Sabokrou.