240 likes | 356 Views
Multi-Task Learning for Boosting with Application to Web Search Ranking Olivier Chapelle et al. Presenter: Wei Cheng. Outline. Motivation Backgrounds Algorithm From svm to boosting using L1 regularization Є -boosting for optimization Overall algorithm Evaluation
E N D
Multi-Task Learning for Boosting with Application to Web Search RankingOlivier Chapelle et al. Presenter: Wei Cheng
Outline • Motivation • Backgrounds • Algorithm • From svm to boosting using L1 regularization • Є-boosting for optimization • Overall algorithm • Evaluation • Overall review and new research points discussion
Motivation Domain specific engine is better! e.g. ‘gelivable’ (very useful) Different/same search engine(s) for different countries?
Motivation • Should we train ranking model separately? • Corps in some domains might be too small to train a good model • Solution: Multi-task learning
Backgrounds SinnoJialin Pan and Qiang Yang, TKDE’2010
Backgrounds SinnoJialin Pan and Qiang Yang, TKDE’2010
Backgrounds • Why transfer learning works? SinnoJialin Pan et al. At WWW’10
Backgrounds • Why transfer learning works?(continue)
Backgrounds • Why transfer learning works?(continue)
Backgrounds traditional learning Input: LearnerA LearnerB Target: Dog/human Girl/boy
Backgrounds Multi-task learning Input: Joint Learning Task Target: Dog/human Girl/boy
Algorithm • The algorithm aims at designing an algorithm based on gradient boosted decision trees • Inspired by svm based multi-task solution and boosting-trick. • Using Є-boosting for optimization
Algorithm • From svm to boosting using L1 regularization • Previous svm based multi-task learning:
Algorithm • Svm(kernel-trick)---boosting (boosting trick) Pick set of non-linear functions(e.g., decision trees, regression trees,….) Apply every single function to each data point H Xф(X) |H|=J
Є-boosting for optimization • Using L1 regulization Using Є-boosting
Evaluation • Datasets
Overall review and new research point discussion • Contributions: • Propose a novel multi-task learning method based on gradient boosted decision tree, which is useful for web-reranking applications. (e.g., personalized search). • Have a thorough evaluation on we-scale datasets. • New research points: • Negative transfer: • Effective grouping: flexible domain adaptation
Q&A Thanks!