深度学习论文: FUN-AD: Fully Unsupervised Learning for Anomaly Detection with Noisy Training Data

深度学习论文: FUN-AD: Fully Unsupervised Learning for Anomaly Detection with Noisy Training Data
FUN-AD: Fully Unsupervised Learning for Anomaly Detection with Noisy Training Data
PDF: https://arxiv.org/pdf/2411.16110
PyTorch代码: https://github.com/shanglianlm0525/CvPytorch
PyTorch代码: https://github.com/shanglianlm0525/PyTorch-Networks

1 概述

尽管异常检测领域的主流研究方向倾向于采用一类分类方法，但在实际的工业生产环境中，训练数据往往因标注错误或新产品、翻新品的标签缺失而遭受噪声污染。针对这些挑战，本文创新性地