It is known that RankSVM can optimize area under the ROC curve (AUC) for binary classification by maximizing the margin between the positive class and the negative class. Since the objective function of Siamese Network for rank learning is the same as RankSVM, Siamese Network can also optimize AUC for binary classification. This paper proposes a method for binary classification by combining Siamese Network for rank learning with logistic regression. The effectiveness is investigated by comparing the AUC scores of the proposed method with the standard Convolutional Neural Network. Then the proposed method is extended to multi-class classification problem by using Siamese Network and multinominal logistic regression. We extend the proposed binary classifier to multi-class classification by using one-vs-others approach.
CITATION STYLE
Oki, H., Miyao, J., & Kurita, T. (2019). Siamese Network for Classification with Optimization of AUC. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11954 LNCS, pp. 315–327). Springer. https://doi.org/10.1007/978-3-030-36711-4_27
Mendeley helps you to discover research relevant for your work.