Machine Learning
Nonlinear component analysis as a kernel eigenvalue problem
Neural Computation
Learning Decision Trees Using the Area Under the ROC Curve
ICML '02 Proceedings of the Nineteenth International Conference on Machine Learning
An efficient boosting algorithm for combining preferences
The Journal of Machine Learning Research
Data mining in metric space: an empirical analysis of supervised learning performance criteria
Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining
Optimising area under the ROC curve using gradient descent
ICML '04 Proceedings of the twenty-first international conference on Machine learning
Generalization Bounds for the Area Under the ROC Curve
The Journal of Machine Learning Research
A support vector method for multivariate performance measures
ICML '05 Proceedings of the 22nd international conference on Machine learning
Stability and generalization of bipartite ranking algorithms
COLT'05 Proceedings of the 18th annual conference on Learning Theory
Margin-Based ranking meets boosting in the middle
COLT'05 Proceedings of the 18th annual conference on Learning Theory
Active Sampling for Rank Learning via Optimizing the Area under the ROC Curve
ECIR '09 Proceedings of the 31th European Conference on IR Research on Advances in Information Retrieval
Training and testing of recommender systems on data missing not at random
Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining
Rule-based active sampling for learning to rank
ECML PKDD'11 Proceedings of the 2011 European conference on Machine learning and knowledge discovery in databases - Volume Part III
Information Sciences: an International Journal
| Hi-index | 0.00 |
In ranking as well as in classification problems, the Area under the ROC Curve (AUC), or the equivalent Wilcoxon-Mann-Whitney statistic, has recently attracted a lot of attention. We show that the AUC can be lower bounded based on the hinge-rank-loss, which simply is the rank-version of the standard (parametric) hinge loss. This bound is asymptotically tight. Our experiments indicate that optimizing the (standard) hinge loss typically is an accurate approximation to optimizing the hinge rank loss, especially when using affine transformations of the data, like e.g. in ellipsoidal machines. This explains for the first time why standard training of support vector machines approximately maximizes the AUC, which has indeed been observed in many experiments in the literature.