English

DOC3-Deep One Class Classification using Contradictions

Machine Learning 2022-05-24 v2 Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning

Abstract

This paper introduces the notion of learning from contradictions (a.k.a Universum learning) for deep one class classification problems. We formalize this notion for the widely adopted one class large-margin loss, and propose the Deep One Class Classification using Contradictions (DOC3) algorithm. We show that learning from contradictions incurs lower generalization error by comparing the Empirical Rademacher Complexity (ERC) of DOC3 against its traditional inductive learning counterpart. Our empirical results demonstrate the efficacy of DOC3 compared to popular baseline algorithms on several real-life data sets.

Keywords

Cite

@article{arxiv.2105.07636,
  title  = {DOC3-Deep One Class Classification using Contradictions},
  author = {Sauptik Dhar and Bernardo Gonzalez Torres},
  journal= {arXiv preprint arXiv:2105.07636},
  year   = {2022}
}

Comments

Deep Learning, Anomaly Detection, Visual Inspection, Learning from Contradictions, Disjoint Auxiliary, Outlier Exposure, MVTec-AD

R2 v1 2026-06-24T02:10:02.494Z