On the Detection of Anomalous or Out-Of-Distribution Data in Vision Models Using Statistical Techniques

arXiv:2403.15497v1 Announce Type: cross
Abstract: Out-of-distribution data and anomalous inputs are vulnerabilities of machine learning systems today, often causing systems to make incorrect predictions. The diverse range of data on which these models are used makes detecting atypical inputs a difficult and important task. We assess a tool, Benford's law, as a method used to quantify the difference between real and corrupted inputs. We believe that in many settings, it could function as a filter for anomalous data points and …

abstract arxiv cs.cv cs.lg data detection distribution diverse inputs learning systems machine machine learning predictions statistical systems type vision vision models vulnerabilities