Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/107297
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | On the importance of pair-wise feature correlations for image classification |
Author: | McDonnell, M. McKilliam, R. De Chazal, P. |
Citation: | Proceedings of International Joint Conference on Neural Networks, 2016, vol.2016-October, pp.2290-2297 |
Publisher: | IEEE |
Publisher Place: | http://www.ijcnn.org |
Issue Date: | 2016 |
Series/Report no.: | IEEE International Joint Conference on Neural Networks (IJCNN) |
ISBN: | 9781509006199 |
ISSN: | 2161-4393 |
Conference Name: | International Joint Conference on Neural Networks (IJCNN 2016) (24 Jul 2016 - 29 Jul 2016 : Vancouver, CANADA) |
Statement of Responsibility: | Mark D. McDonnell, Robby G. McKilliam, Philip de Chazal |
Abstract: | We show that simple linear classification of pairwise products of convolutional features achieves near state-of-the-art performance on some standard labelled image databases. Specifically, we found test classification error rates on the MNIST handwritten digits image database of under 0.5%, and achieved under 19% and under 44% error rates on the CIFAR-10 and CIFAR-100 RGB image databases. Since the number of weights in such a classifier grows with the square of the number of features, we discuss how implementation of such a pair-wise products classifier can be achieved in an SLFN architecture where the hidden unit function is the simple quadratic nonlinearity: we can this a Quadratic Neural Network (QNN). We compare this method to setting the input weights in a QNN randomly, and find optimal performance can be achieved provided the hidden layer is sufficiently large. This analysis provides insight on why `extreme-learning machines' can achieve classification performance equal to or better than the use of backpropagation training. |
Description: | IJCNN 2016 (as part of WCCI 2016) |
Rights: | ©2016 IEEE |
DOI: | 10.1109/IJCNN.2016.7727483 |
Grant ID: | http://purl.org/au-research/grants/arc/DP1093425 http://purl.org/au-research/grants/arc/FT110101098 |
Published version: | http://dx.doi.org/10.1109/ijcnn.2016.7727483 |
Appears in Collections: | Aurora harvest 3 Electrical and Electronic Engineering publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.