Skoltech researcher Bogdan Kirillov and Maxim Panov (Technology Innovation Institute, Abu Dhabi, United Arab Emirates) have published a paper “Measuring internal inequality in capsule networks for supervised anomaly detection” in Scientific Reports. The study is focused on the use of income inequality metrics such as Gini or Palma coefficients as a tool to identify anomalies via capsule networks. The researchers explore how the interplay between primary and class capsules gives rise to differences in behavior regarding anomalous and normal input which can be exploited to detect anomalies. They derive several criteria for capsule networks and apply them to a number of Computer Vision benchmark datasets, as well as to the dataset of skin lesion images and the dataset of CRISPR-Cas9 off-target pairs. Full text of the paper is available here.