Probabilistic retrieval based on staged logistic regression
SIGIR '92 Proceedings of the 15th annual international ACM SIGIR conference on Research and development in information retrieval
Theoretical Views of Boosting and Applications
ALT '99 Proceedings of the 10th International Conference on Algorithmic Learning Theory
An efficient boosting algorithm for combining preferences
The Journal of Machine Learning Research
Learning to rank using gradient descent
ICML '05 Proceedings of the 22nd international conference on Machine learning
Learning to rank: from pairwise approach to listwise approach
Proceedings of the 24th international conference on Machine learning
Ranking with multiple hyperplanes
SIGIR '07 Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval
FRank: a ranking method with fidelity loss
SIGIR '07 Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval
SoftRank: optimizing non-smooth rank metrics
WSDM '08 Proceedings of the 2008 International Conference on Web Search and Data Mining
Hi-index | 0.00 |
Over the last decade learning to rank (L2R) has gained a lot of attention and many algorithms have been proposed. One of the most successful approach is to build an algorithm following the ensemble principle. Boosting is the key representative of this approach. However, even boosting isn't effective when used to increase the performance of individually strong algorithms, scenario when we want to blend already successful L2R algorithms in order to gain an additional benefit. To address this problem we propose a novel algorithm, based on a theory of nonlinear monotonic ensembles, which is able to blend strong base rankers effectively. Specifically, we provide the concept of defect of a set of algorithms that allows to deduce a popular pairwise approach in strict mathematical terms. Using the concept of defect, we formulate an optimization problem and propose a sound method of its solution. Finally, we conduct experiments with real data which shows the effectiveness of the proposed approach.