Ranking, Boosting, and Model Adaptation
Chris J. C. Burges, Krysta M. Svore, Qiang Wu, Jianfeng Gao · 2008
We present a new ranking algorithm that combines the strengths of two previous methods: boosted tree classification, and LambdaR ank, which has been shown to be empirically optimal for a widely used information retrieval measure. The algorithm is based on boosted regression trees, although the ideas apply to any weak learners, and it is significantly fast er in both train and test phases than the state of the art, for comparable accuracy. We also show how to find the optimal linear combination for any two ran kers, and we use this method to solve the line search problem exactly during boosting. In addition, we show that starting with a previously tra ined model, and boosting using its residuals, furnishes an effective techn ique for model adaptation, and we give results for a particularly pressing prob lem in Web Search - training rankers for markets for which only small amounts of labeled data are available, given a ranker trained on much more data from a larger market.