Fault Detection of LLE Compound Statistic Based on Weighted K-nearest Neighbor Selection

Gu, Haoyu; Qian, Ping; Wang, Li; Cao, Kaitian

doi:10.1109/cac48633.2019.8997239

Cited by 1 publication

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typically, anomaly detection assumes that an instance is considered anomalous if its pattern significantly deviates from the majority. These approaches are commonly implemented using clustering-based methods [6], neighborhood-based methods [11], or reconstruction-based methods [53]. However, with the prevalence of multi-view data in today's world, where instances are described by multiple views or modalities, such as different news organizations reporting the same news or an image being encoded by various features, traditional anomaly detection methods that focus on single-view data are insufficient.…”

Section: Introductionmentioning

confidence: 99%

Debunking Free Fusion Myth: Online Multi-view Anomaly Detection with Disentangled Product-of-Experts Modeling

Wang,

Cheng,

Sun

et al. 2023

Proceedings of the 31st ACM International Conference on Multimedia

View full text Add to dashboard Cite

Multi-view or even multi-modal data is appealing yet challenging for real-world applications. Detecting anomalies in multi-view data is a prominent recent research topic. However, most of the existing methods 1) are only suitable for two views or type-specific anomalies, 2) suffer from the issue of fusion disentanglement, and 3) do not support online detection after model deployment. To address these challenges, our main ideas in this paper are three-fold: multiview learning, disentangled representation learning, and generative model. To this end, we propose dPoE, a novel multi-view variational autoencoder model that involves (1) a Product-of-Experts (PoE) layer in tackling multi-view data, (2) a Total Correction (TC) discriminator in disentangling view-common and view-specific representations, and (3) a joint loss function in wrapping up all components. In addition, we devise theoretical information bounds to control both view-common and view-specific representations. Extensive experiments on six real-world datasets demonstrate that the proposed dPoE outperforms baselines markedly.

show abstract

Section: Introductionmentioning

confidence: 99%