Fine-tuning deep neural networks in continuous learning scenarios
The revival of deep neural networks and the availability of ImageNet laid the foundation for
recent success in highly complex recognition tasks. However, ImageNet does not cover all …
recent success in highly complex recognition tasks. However, ImageNet does not cover all …
Weighted support vector data description based on chaotic bat algorithm
J Hamidzadeh, R Sadeghi, N Namaei - Applied Soft Computing, 2017 - Elsevier
Abstract Support Vector Data Description (SVDD) is a support vector based learning
algorithm for anomaly detection. In this method, the target is to form a boundary around the …
algorithm for anomaly detection. In this method, the target is to form a boundary around the …
A fast algorithm of convex hull vertices selection for online classification
Reducing samples through convex hull vertices selection (CHVS) within each class is an
important and effective method for online classification problems, since the classifier can be …
important and effective method for online classification problems, since the classifier can be …
Superpixel-based online wagging one-class ensemble for feature selection in foreground/background separation
In the last decades, researchers in the field of Background Subtraction (BS) have developed
methods to handle the different type of challenges. However, at the present time, no …
methods to handle the different type of challenges. However, at the present time, no …
An efficient semi-supervised SVM for anomaly detection
P Montague, J Kim - 2017 International Joint Conference on …, 2017 - ieeexplore.ieee.org
Semi-Supervised Support Vector Machines (S3VMs) have been proposed to deal with the
proliferation of partially labelled data available in many large-scale complex systems. Since …
proliferation of partially labelled data available in many large-scale complex systems. Since …
Forgetting of unused classes in missing data environment using automatically generated data: Application to on-line handwritten gesture command recognition
In this paper we exploit the use of synthetic data for on-line handwritten gesture commands
recognition with an emphasis on the problem of forgetting unused classes. For on-line …
recognition with an emphasis on the problem of forgetting unused classes. For on-line …
Outlier detection by consistent data selection method
U Porwal, S Mukund - arXiv preprint arXiv:1712.04129, 2017 - arxiv.org
Often the challenge associated with tasks like fraud and spam detection [1] is the lack of all
likely patterns needed to train suitable supervised learning models. In order to overcome this …
likely patterns needed to train suitable supervised learning models. In order to overcome this …
Automated Optokinetic Nystagmus Detection for Use with Young Subjects
M Sangi - 2017 - researchspace.auckland.ac.nz
There are currently no objective tests of visual function for young children and the tests that
do exist are subjective, can only be applied reliably in children over 3 years of age and …
do exist are subjective, can only be applied reliably in children over 3 years of age and …
Two kinds of targets on-line classification based on incremental SVDD
B Lei, H Xiao, Y Guo - 2017 IEEE 3rd Information Technology …, 2017 - ieeexplore.ieee.org
In this paper, a novel classifier for classification problems, based on increment support
vector data description, is proposed. The proposed method is the expand version of …
vector data description, is proposed. The proposed method is the expand version of …
Online learning with regularized kernel for one-class classification
This paper presents an online learning with regularized kernel based one-class extreme
learning machine (ELM) classifier and is referred as online RK-OC-ELM. The baseline …
learning machine (ELM) classifier and is referred as online RK-OC-ELM. The baseline …