|
Title:
|
Minimax support vector machines |
|
Author:
|
Davenport, Mark A.; Baraniuk, Richard G.; Scott, Clayton D.
|
|
Type:
|
Article |
|
Citation:
|
M. A. Davenport, R. G. Baraniuk and C. D. Scott, "Minimax support vector machines," IEEE Workshop on Statistical Signal Processing (SSP), 2007. |
|
Abstract:
|
We study the problem of designing support vector machine (SVM)
classifiers that minimize the maximum of the false alarm and miss
rates. This is a natural classification setting in the absence of prior
information regarding the relative costs of the two types of errors
or true frequency of the two classes in nature. Examining two approaches
– one based on shifting the offset of a conventionally trained
SVM, the other based on the introduction of class-specific weights –
we find that when proper care is taken in selecting the weights, the
latter approach significantly outperforms the strategy of shifting the
offset. We also find that the magnitude of this improvement depends
chiefly on the accuracy of the error estimation step of the training
procedure. Furthermore, comparison with the minimax probability
machine (MPM) illustrates that our SVM approach can outperform
the MPM even when the MPM parameters are set by an oracle. |
|
Date Published:
|
2007-08-01 |
