Kernel PCA for novelty detection

By Heiko Hoffmann. 2007

808 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Molecular Biology

Citations in the corpus

Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications (2021, neutral): “, centroid-based [119], PCA-based [64], kernelbased [141], or combinations of them [37], [72], [164])”
Complete Random Forest Based Class Noise Filtering Learning for Improving the Generalizability of Classifiers (2019, neutral): “As class noise and anomaly have some similarity, some distance based anomaly detection techniques can be used for class noise detection [33], [34].”
OCGAN: One-Class Novelty Detection Using GANs With Constrained Latent Representations (2019, methodology): “The earliest methods in one-class novelty detection used Principal Component Analysis (PCA) [2] and its kernel extension [5] to find a subspace that best describes the given concept.”
Outlier Analysis (2013, methodology): “Numerous methods have been designed for this case such as single-class SVMs [397, 91], minimax probability machines [282], kernel-based PCA methods [383], direct density ratio estimation [214], and extreme value analysis [220].”
Local Novelty Detection in Multi-class Recognition Problems (2015, methodology): “Further one-class classiﬁcation techniques range from prototype based approaches [1, 2], Parzen density estimation [4], convex hull algorithms [7], as well as one-class random forests [9] to kernel PCA [16] and kernel Fisher dis-criminants [30].”
Face Recognition Systems: A Survey (2020, methodology): “KPCA has been successfully used for novelty detection [72] or for speech recognition [62].”
Learning Sets with Separating Kernels (2012, methodology): “A kernel method based on kernel principal component analysis is presented in [35] and is essentially a special case of our framework.”
A Unifying Review of Deep and Shallow Anomaly Detection (2020, methodology): “However, it is clear from an inspection of the explanation that KDE is not looking at the true cause of the anomaly and is looking instead at the vertical stripes present everywhere in the input image.”
Anomaly Detection Using Autoencoders with Nonlinear Dimensionality Reduction (2014, methodology): “Kernel PCA [5] performs nonlinear mapping to the highdimensional feature space by a kernel function, and then employ linear PCA in the feature space.”
Distil : A Mixed-Initiative Model Discovery System for Subject Matter Experts ( Demo ) (2018, background): “In our system, PUNK (New Knowledge, 2017) applies a PCA-based ranking algorithm, where feature weightings of the first principal component or the average weighting across the top N principal components derive feature interestingness.”
Unsupervised Deep Anomaly Detection for Multi-Sensor Time-Series Signals (2021, methodology): “In order to extensively evaluate the performance of the proposed CAE-M approach, we compare it with several traditional and deep anomaly detection methods: (1) Kernel principal component analysis (KPCA) [28], which is a non-linear extension of PCA commonly used”
Dimensionality Reduction: A Comparative Review (2008, background): “, face recognition [63], speech recognition [75], and novelty detection [55].”
Robust Anomaly Detection and Backdoor Attack Detection Via Differential Privacy (2019, methodology): “Traditional methods such as clustering ([30]) and principal component analysis (PCA) ([31, 32]) have been frequently used.”