DOC3-Deep One Class Classification using Contradictions

원문 초록

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.

함께 살펴볼 논문

공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.

논문 ID: document/1082898공통 태그 Anomaly Detection, Classification 등록 2024-10-26

AI-Enhanced Pediatric Pneumonia Detection: A CNN-Based Approach Using Data Augmentation and Generative Adversarial Networks (GANs)

arXiv:2507.07435공통 태그 Anomaly Detection공식 코드 · Papers with Code 등록 2025-07-10

Towards High-Resolution 3D Anomaly Detection: A Scalable Dataset and Real-Time Framework for Subtle Industrial Defects

토론

이 논문과 연결된 포럼 글이 아직 없습니다.

포럼에서 토론 찾기

DOC3-Deep One Class Classification using Contradictions

원문 초록

함께 살펴볼 논문

Attention to Patterns is all you need for Insider threat detection

Anomaly Detection Using Normalizing Flow-Based Density Estimation and Synthetic Defect Classification

Anomaly Detection in File Fragment Classification of Image File Formats

Efficient Calisthenics Skills Classification through Foreground Instance Selection and Depth Estimation

AI-Enhanced Pediatric Pneumonia Detection: A CNN-Based Approach Using Data Augmentation and Generative Adversarial Networks (GANs)

Towards High-Resolution 3D Anomaly Detection: A Scalable Dataset and Real-Time Framework for Subtle Industrial Defects

토론