Title :
Explicit feature mapping via multi-layer perceptron and its application to Mine-Like Objects detection
Author :
Hang Shao ; Japkowicz, Nathalie
Author_Institution :
Sch. of Electr. Eng. & Comput. Sci., Univ. of Ottawa, Ottawa, ON, Canada
Abstract :
In this paper, a novel learning method is introduced that borrows simultaneously from the principles of kernel methods and multi-layer perceptron. Specifically, the method implements the feature mapping idea of kernel methods into a multi-layer perceptron. Unlike in kernel learning where the feature space is usually invisible and inaccessible, the multilayer perceptron based mapping is explicit. Therefore, the proposed model can be learned directly in feature space. Together with the inherent sparse representation, the proposed approach will thus be much faster and easier to train even in the event of a large network size. The proposed approach is applied in the context of an Autonomous Underwater Vehicle Mine-Like Objects detection task. The results show that the proposed approach is able to improve upon the generalization performance of neural network based methods. Its prediction results are also close to or better than those obtained by kernel machines. Its learning and classification speed is shown to far surpass those of kernel machines. These results are confirmed on a number of experiments involving benchmarking UCI domains.
Keywords :
feature extraction; image classification; image representation; learning (artificial intelligence); multilayer perceptrons; object detection; autonomous underwater vehicle; classification speed; explicit feature mapping; feature space; generalization performance; kernel machines; kernel methods; learning method; mine-like objects detection; multilayer perceptron based mapping; network size; neural network based methods; sparse representation; Buildings; Kernel; Learning systems; Neurons; Sonar; Testing; Training; Feature mapping; Kernel learning; Machine learning; Neural networks; Target detection;
Conference_Titel :
Neural Networks (IJCNN), 2014 International Joint Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4799-6627-1
DOI :
10.1109/IJCNN.2014.6889450