Name
Title | ||
|---|---|---|
A New Cost Function For Binary Classification Problems Based On The Distributions Of The Soft Output For Each Class |
Abstract | ||
|---|---|---|
This paper proposes a new cost function for supervised training of neural networks in binary classification applications. This cost function aims at reducing the probability of classification error by reducing the overlap between distributions of the soft output for each class. The non parametric Parzen window method, with Gaussian kernels, is used to estimate the distributions from the training data set. The cost function has been implemented in a GRBF neural network and has been tested in a motion detection application from low resolution infrared images, showing some advantages with respect to the conventional mean squared error cost function and also with respect to the support vector machine, a reference binary classifier. |
Year | DOI | Venue |
|---|---|---|
2007 | 10.1109/IJCNN.2007.4371174 | 2007 IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-6 |
Keywords | Field | DocType |
support vector machine,nonparametric statistics,gaussian processes,support vector machines,cost function,neural networks,mean square error,gaussian kernel,learning artificial intelligence,binary classification,neural network,mean squared error,low resolution | Binary classification,Pattern recognition,Computer science,Support vector machine,Mean squared error,Nonparametric statistics,Gaussian,Artificial intelligence,Gaussian process,Artificial neural network,Machine learning,Kernel density estimation | Conference |
ISSN | Citations | PageRank |
1098-7576 | 0 | 0.34 |
References | Authors | |
3 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Marcelino Lázaro | 1 | 78 | 11.34 |
| José M. Leiva-Murillo | 2 | 31 | 3.85 |
| Antonio Artés-Rodríguez | 3 | 206 | 34.76 |
| Aníbal R. Figueiras-Vidal | 4 | 467 | 38.03 |