A View on Out-of-Distribution Identification from a Statistical Testing Theory Perspective

arXiv:2405.03052v1 Announce Type: new
Abstract: We study the problem of efficiently detecting Out-of-Distribution (OOD) samples at test time in supervised and unsupervised learning contexts. While ML models are typically trained under the assumption that training and test data stem from the same distribution, this is often not the case in realistic settings, thus reliably detecting distribution shifts is crucial at deployment. We re-formulate the OOD problem under the lenses of statistical testing and then discuss conditions that render the OOD …

abstract arxiv case cs.lg data distribution identification ml models perspective samples statistical stem study test testing theory training type unsupervised unsupervised learning view while