Title | Neural network classification: Maximizing zero-error density |
Publication Type | Conference Paper |
Year of Publication | 2005 |
Authors | Silva, LM, Alexandre, LA, De Sá, JM |
Editor | S., S, M., S, C., A, P., P |
Conference Name | Lecture Notes in Computer ScienceLect. Notes Comput. Sci. |
Date Published | 2005/// |
Conference Location | Bath |
ISBN Number | 03029743 (ISSN) |
Keywords | Algorithms, Backpropagation, Backpropagation algorithm, Classification (of information), Costs, Error analysis, Function evaluation, Learning systems, Mean square error, Neural network classification, Neural networks, Zero-error density |
Abstract | We propose a new cost function for neural network classification: the error density at the origin. This method provides a simple objective function that can be easily plugged in the usual backpropagation algorithm, giving a simple and efficient learning scheme. Experimental work shows the effectiveness and superiority of the proposed method when compared to the usual mean square error criteria in four well known datasets. © Springer-Verlag Berlin Heidelberg 2005. |
URL | http://www.scopus.com/inward/record.url?eid=2-s2.0-27244450344&partnerID=40&md5=73ee3c0c073fb3bc7bc85013bf7dcce6 |