Incremental Support Vector Learning: Analysis, Implementation and Applications

By Pavel Laskov, Christian Gehl, Stefan Krüger, K. Müller. 2006

379 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Security analysis of online centroid anomaly detection (2010, background): “Obviously, re-training the model from scratch every time is computationally infeasible; however, incorporation of new data points and the removal of irrelevant ones can be done with acceptable effort (Laskov et al., 2006).”
Towards Open World Recognition (2014, background): “The update process is extremely expensive (quadratic in the number of training examples learned [19]) and depends heavily on the number of support vectors stored for performing updates [19].”
FIRM: An Intelligent Fine-Grained Resource Management Framework for SLO-Oriented Microservices (2020, methodology): “Then, for each CP, we calculate feature statistics and feed them into an incremental SVM classifier [29,58] implemented using stochastic gradient descent optimization and RBF kernel approximation by scikit-learn libraries [91].”
Support vector machines (2008, unknown): “3, with the parameters of the abstract formulation (56) defined as: c = diag(K), a = y and b = −1 (Laskov et al., 2006).”
The Extreme Value Machine (2015, background): “ing problem, they are computationally expensive [1], [2], [3], [4], or provide little to no characterization of the statistical distribution of the data [5], [6], [7], [8].”
Incremental Support Vector Learning for Ordinal Regression (2015, methodology): “Laskov et al. [18] implemented an accurate incremental one-class SVM algorithm.”
Incremental learning for ν-Support Vector Regression (2015, methodology): “Based on the incremental and decre-mental ν -SVR algorithms, we can implement leave-one-out cross validation (Weston, 1999) and the learning with limited memory (Laskov et al., 2006) efficiently.”
Incremental robot learning of new objects with fixed update time (2016, unknown): “Such formulation typically allows to compute the current model in closed form (or with few operations independent of the number of examples) as a combination of the previous model and the new observed example [5], [18].”