Title :
Incremental probabilistic classification vector machine with linear costs
Author :
F.-M. Schleif;H. Chen;P. Tino
Author_Institution :
University of Birmingham, School of Computer Science, B15 2TT, UK
fDate :
7/1/2015 12:00:00 AM
Abstract :
The probabilistic classification vector machine is a very effective and generic probabilistic and sparse classifier. A recently published incremental version improved the runtime complexity to quadratic costs. We derive the Nyström approximation for asymmetric matrices to obtain linear runtime and memory complexity for the incremental probabilistic classification vector machine while keeping similar prediction performance.
Keywords :
"Xenon","Support vector machines"
Conference_Titel :
Neural Networks (IJCNN), 2015 International Joint Conference on
Electronic_ISBN :
2161-4407
DOI :
10.1109/IJCNN.2015.7280377