ABSTRACT
This paper studies data optimization for Learning to Rank (LtR), by dropping training labels to increase ranking accuracy. Our work is inspired by data dropout, showing some training data do not positively influence learning and are better dropped out, despite a common belief that a larger training dataset is beneficial. Our main contribution is to extend this intuition for noisy- and semi- supervised LtR scenarios: some human annotations can be noisy or out-of-date, and so are machine-generated pseudo-labels in semi- supervised scenarios. Dropping out such unreliable labels would contribute to both scenarios. State-of-the-arts propose Influence Function (IF) for estimating how each training instance affects learn- ing, and we identify and overcome two challenges specific to LtR. 1) Non-convex ranking functions violate the assumptions required for the robustness of IF estimation. 2) The pairwise learning of LtR incurs quadratic estimation overhead. Our technical contributions are addressing these challenges: First, we revise estimation and data optimization to accommodate reduced reliability; Second, we devise a group-wise estimation, reducing cost yet keeping accuracy high. We validate the effectiveness of our approach in a wide range of ad-hoc information retrieval benchmarks and real-life search engine datasets in both noisy- and semi-supervised scenarios.
Supplemental Material
- Qingyao Ai, Keping Bi, Cheng Luo, Jiafeng Guo, and W Bruce Croft. 2018. Unbiased learning to rank with unbiased propensity estimation. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval. 385--394.Google ScholarDigital Library
- Samyadeep Basu, Xuchen You, and Soheil Feizi. 2019. Second-Order Group Influence Functions for Black-Box Predictions. arXiv preprint arXiv:1911.00418 (2019).Google Scholar
- Sebastian Bruch. 2019. An Alternative Cross Entropy Loss for Learning-to-Rank. arXiv preprint arXiv:1911.09798 (2019).Google Scholar
- Christopher Burges, Tal Shaked, Erin Renshaw, Ari Lazier, Matt Deeds, Nicole Hamilton, and Gregory N Hullender. 2005. Learning to rank using gradient descent. In Proceedings of the 22nd International Conference on Machine learning (ICML-05). 89--96.Google ScholarDigital Library
- Christopher JC Burges. 2010. From ranknet to lambdarank to lambdamart: An overview. Learning 11, 23--581 (2010), 81.Google Scholar
- Olivier Chapelle and Ya Zhang. 2009. A dynamic bayesian network click model for web search ranking. In Proceedings of the 18th international conference on World wide web. 1--10.Google ScholarDigital Library
- Charles L Clarke, Nick Craswell, and Ian Soboroff. 2009. Overview of the trec 2009 web track. Technical Report. WATERLOO UNIV (ONTARIO).Google Scholar
- Djork-Arné Clevert, Thomas Unterthiner, and Sepp Hochreiter. 2015. Fast and accurate deep network learning by exponential linear units (elus). arXiv preprint arXiv:1511.07289 (2015).Google Scholar
- Faïza Dammak, Hager Kammoun, and Abdelmajid Ben Hamadou. 2017. Improving pairwise learning to rank algorithms for document retrieval. In 2017 IEEE Symposium Series on Computational Intelligence (SSCI). IEEE, 1--8.Google ScholarCross Ref
- Dany Haddad and Joydeep Ghosh. 2019. Learning More From Less: Towards Strengthening Weak Supervision for Ad-Hoc Retrieval. In Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. ACM, 857--860.Google ScholarDigital Library
- Thorsten Joachims. 2002. Optimizing search engines using clickthrough data. In Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 133--142.Google ScholarDigital Library
- Pang Wei Koh and Percy Liang. 2017. Understanding black-box predictions via influence functions. In Proceedings of the 34th International Conference on Machine Learning-Volume 70. JMLR. org, 1885--1894.Google ScholarDigital Library
- Pang Wei W Koh, Kai-Siang Ang, Hubert Teo, and Percy S Liang. 2019. On the accuracy of influence functions for measuring group effects. In Advances in Neural Information Processing Systems. 5254--5264.Google Scholar
- Zhigao Miao, Juan Wang, Aimin Zhou, and Ke Tang. 2015. Regularized boost for semi-supervised ranking. In Proceedings of the 18th Asia Pacific Symposium on Intelligent and Evolutionary Systems, Volume 1. Springer, 643--651.Google ScholarCross Ref
- Dae Hoon Park and Yi Chang. 2019. Adversarial Sampling and Training for Semi-Supervised Information Retrieval. In TheWorld WideWeb Conference. ACM, 1443--1453.Google Scholar
- Tao Qin and Tie-Yan Liu. 2013. Introducing LETOR 4.0 Datasets. CoRRabs/1306.2597 (2013). http://arxiv.org/abs/1306.2597Google Scholar
- Tao Qin and Tie-Yan Liu. 2013. Introducing LETOR 4.0 datasets. arXiv preprint arXiv:1306.2597 (2013).Google Scholar
- Jun Wang, Lantao Yu, Weinan Zhang, Yu Gong, Yinghui Xu, Benyou Wang, Peng Zhang, and Dell Zhang. 2017. Irgan: A minimax game for unifying generative and discriminative information retrieval models. In Proceedings of the 40th International ACM SIGIR conference on Research and Development in Information Retrieval. ACM, 515--524.Google ScholarDigital Library
- Tianyang Wang, Jun Huan, and Bo Li. 2018. Data dropout: Optimizing training data for convolutional neural networks. In 2018 IEEE 30th International Conference on Tools with Artificial Intelligence (ICTAI). IEEE, 39--46.Google ScholarCross Ref
- XuanhuiWang, Cheng Li, Nadav Golbandi, Michael Bendersky, and Marc Najork. 2018. The lambdaloss framework for ranking metric optimization. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management. 1313--1322.Google Scholar
- Zifeng Wang, Hong Zhu, Zhenhua Dong, Xiuqiang He, and Shao-Lun Huang. 2019. Less Is Better: Unweighted Data Subsampling via Influence Function. arXiv preprint arXiv:1912.01321 (2019).Google Scholar
- Jingfang Xu, Chuanliang Chen, Gu Xu, Hang Li, and Elbio Renato Torres Abib. 2010. Improving quality of training data for learning to rank using click-through data. In Proceedings of the third ACM international conference on Web search and data mining. 171--180.Google ScholarDigital Library
- Minjie Xu and Gary Kazantsev. 2019. Understanding Goal-Oriented Active Learning via Influence Functions. arXiv preprint arXiv:1905.13183 (2019).Google Scholar
- Shipeng Yu, Deng Cai, Ji-Rong Wen, and Wei-Ying Ma. 2003. Improving pseudorelevance feedback in web information retrieval using web page segmentation. In Proceedings of the 12th international conference on World Wide Web. ACM, 11--18.Google ScholarDigital Library
Index Terms
- Training Data Optimization for Pairwise Learning to Rank
Recommendations
Training query filtering for semi-supervised learning to rank with pseudo labels
Semi-supervised learning is a machine learning paradigm that can be applied to create pseudo labels from unlabeled data for learning a ranking model, when there is only limited or no training examples available. However, the effectiveness of semi-...
Inductive Semi-supervised Multi-Label Learning with Co-Training
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningIn multi-label learning, each training example is associated with multiple class labels and the task is to learn a mapping from the feature space to the power set of label space. It is generally demanding and time-consuming to obtain labels for training ...
Learning to Rank from Noisy Data
Learning to rank, which learns the ranking function from training data, has become an emerging research area in information retrieval and machine learning. Most existing work on learning to rank assumes that the training data is clean, which is not ...
Comments