TY - GEN
T1 - Incremental One-Class Learning with Bounded Computational Complexity
AU - Sillito, Rowland R.
AU - Fisher, Bob.
PY - 2007
Y1 - 2007
N2 - An incremental one-class learning algorithm is proposed for the purpose of outlier detection. Outliers are identified by estimating - and thresholding - the probability distribution of the training data. In the early stages of training a non-parametric estimate of the training data distribution is obtained using kernel density estimation. Once the number of training examples reaches the maximum computationally feasible limit for kernel density estimation, we treat the kernel density estimate as a maximally-complex Gaussian mixture model, and keep the model complexity constant by merging a pair of components for each new kernel added. This method is shown to outperform a current state-of-the-art incremental one-class learning algorithm (Incremental SVDD [5]) on a variety of datasets, while requiring only an upper limit on model complexity to be specified.
AB - An incremental one-class learning algorithm is proposed for the purpose of outlier detection. Outliers are identified by estimating - and thresholding - the probability distribution of the training data. In the early stages of training a non-parametric estimate of the training data distribution is obtained using kernel density estimation. Once the number of training examples reaches the maximum computationally feasible limit for kernel density estimation, we treat the kernel density estimate as a maximally-complex Gaussian mixture model, and keep the model complexity constant by merging a pair of components for each new kernel added. This method is shown to outperform a current state-of-the-art incremental one-class learning algorithm (Incremental SVDD [5]) on a variety of datasets, while requiring only an upper limit on model complexity to be specified.
U2 - 10.1007/978-3-540-74690-4_7
DO - 10.1007/978-3-540-74690-4_7
M3 - Conference contribution
SN - 978-3-540-74689-8
VL - 4668
T3 - Lecture Notes in Computer Science
SP - 58
EP - 67
BT - Artificial Neural Networks - ICANN 2007
A2 - Marques de Sa, Joaquim
A2 - Alexandre, Luis A
A2 - Duch, Wlodzislaw
A2 - Mandic, Danilo
PB - Springer-Verlag GmbH
ER -