An efficient boosting algorithm for combining preferences
The Journal of Machine Learning Research
Ordering by weighted number of wins gives a good ranking for weighted tournaments
SODA '06 Proceedings of the seventeenth annual ACM-SIAM symposium on Discrete algorithm
SIAM Journal on Discrete Mathematics
Journal of Artificial Intelligence Research
Ranking and scoring using empirical risk minimization
COLT'05 Proceedings of the 18th annual conference on Learning Theory
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
ACM SIGACT News
Aggregating inconsistent information: Ranking and clustering
Journal of the ACM (JACM)
Smooth Boosting for Margin-Based Ranking
ALT '08 Proceedings of the 19th international conference on Algorithmic Learning Theory
Ordering by weighted number of wins gives a good ranking for weighted tournaments
ACM Transactions on Algorithms (TALG)
Reducing position-sensitive subset ranking to classification
Canadian AI'11 Proceedings of the 24th Canadian conference on Advances in artificial intelligence
Approximate reduction from AUC maximization to 1-norm soft margin optimization
ALT'11 Proceedings of the 22nd international conference on Algorithmic learning theory
The K-armed dueling bandits problem
Journal of Computer and System Sciences
Generic subset ranking using binary classifiers
Theoretical Computer Science
Multi-prototype label ranking with novel pairwise-to-total-rank aggregation
IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence
The Journal of Machine Learning Research
Hi-index | 0.00 |
We reduce ranking, as measured by the Area Under the Receiver Operating Characteristic Curve (AUC), to binary classification. The core theorem shows that a binary classification regret of r on the induced binary problem implies an AUC regret of at most 2r. This is a large improvement over approaches such as ordering according to regressed scores, which have a regret transform of r → nr where n is the number of elements.