首页|Zurich University of Applied Sciences Researchers Detail Findings in Machine Lea rning (A Generic Machine Learning Framework for Fully-Unsupervised Anomaly Detec tion with Contaminated Data)
Zurich University of Applied Sciences Researchers Detail Findings in Machine Lea rning (A Generic Machine Learning Framework for Fully-Unsupervised Anomaly Detec tion with Contaminated Data)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News – Investigators discuss new findings in artificial intelligence. According to news reporting originating from Zurich University of Applied Sciences by NewsRx correspondents, research stated, “Anomaly detection ( AD) tasks have been solved using machine learning algorithms in various domains and applications. The great majority of these algorithms use normal data to trai n a residual-based model, and assign anomaly scores to unseen samples based on t heir dissimilarity with the learned normal regime.” Our news editors obtained a quote from the research from Zurich University of Ap plied Sciences: “The underlying assumption of these approaches is that anomaly-f ree data is available for training. This is, however, often not the case in real -world operational settings, where the training data may be contaminated with a certain fraction of abnormal samples. Training with contaminated data, in turn, inevitably leads to a deteriorated AD performance of the residual-based algorith ms. In this paper we introduce a framework for a fully unsupervised refinement o f contaminated training data for AD tasks. The framework is generic and can be a pplied to any residual-based machine learning model. We demonstrate the applicat ion of the framework to two public datasets of multivariate time series machine data from different application fields. We show its clear superiority over the n aive approach of training with contaminated data without refinement.”
Zurich University of Applied SciencesA lgorithmsCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
