Neural Transformation Learning for Deep Anomaly Detection Beyond Images

          Chen Qiu 1 2 Timo Pfrommer 1 Marius Kloft 2 Stephan Mandt 3 Maja Rudolph 1

             Abstract                formations are useful, and it is hard to design these trans-
    Data transformations (e.g. rotations, reections,     formations manually. This paper studies self-supervised
    and cropping) play an important role in self-       anomaly detection for data types beyond images. We de-
    supervised lea ...