Generalized mean for feature extraction in one-class classification problems

• Generalized mean is used for BDA in place of arithmetic mean.
• The proposed method is more robust to outliers than the other alternative methods.
• We also propose a novel method to efficiently maximize the generalized mean.

Biased discriminant analysis (BDA), which extracts discriminative features for one-class classification problems, is sensitive to outliers in negative samples. This study focuses on the drawback of BDA attributed to the objective function based on the arithmetic mean in one-class classification problems, and proposes an objective function based on a generalized mean. A novel method is also presented to effectively maximize the objective function. The experimental results show that the proposed method provides better discriminative features than the BDA and its variants.