Coupled Group Lasso for Web-Scale CTR Prediction in Display Advertising

Ling Yan, Wu-Jun Li, Gui-Rong Xue, Dingyi Han
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(2):802-810, 2014.

Abstract

In display advertising, click through rate(CTR) prediction is the problem of estimating the probability that an advertisement (ad) is clicked when displayed to a user in a specific context. Due to its easy implementation and promising performance, logistic regression(LR) model has been widely used for CTR prediction, especially in industrial systems. However, it is not easy for LR to capture the nonlinear information, such as the conjunction information, from user features and ad features. In this paper, we propose a novel model, called coupled group lasso(CGL), for CTR prediction in display advertising. CGL can seamlessly integrate the conjunction information from user features and ad features for modeling. Furthermore, CGL can automatically eliminate useless features for both users and ads, which may facilitate fast online prediction. Scalability of CGL is ensured through feature hashing and distributed implementation. Experimental results on real-world data sets show that our CGL model can achieve state-of-the-art performance on web-scale CTR prediction tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v32-yan14, title = {Coupled Group Lasso for Web-Scale CTR Prediction in Display Advertising}, author = {Yan, Ling and Li, Wu-Jun and Xue, Gui-Rong and Han, Dingyi}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {802--810}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/yan14.pdf}, url = {https://proceedings.mlr.press/v32/yan14.html}, abstract = {In display advertising, click through rate(CTR) prediction is the problem of estimating the probability that an advertisement (ad) is clicked when displayed to a user in a specific context. Due to its easy implementation and promising performance, logistic regression(LR) model has been widely used for CTR prediction, especially in industrial systems. However, it is not easy for LR to capture the nonlinear information, such as the conjunction information, from user features and ad features. In this paper, we propose a novel model, called coupled group lasso(CGL), for CTR prediction in display advertising. CGL can seamlessly integrate the conjunction information from user features and ad features for modeling. Furthermore, CGL can automatically eliminate useless features for both users and ads, which may facilitate fast online prediction. Scalability of CGL is ensured through feature hashing and distributed implementation. Experimental results on real-world data sets show that our CGL model can achieve state-of-the-art performance on web-scale CTR prediction tasks.} }
Endnote
%0 Conference Paper %T Coupled Group Lasso for Web-Scale CTR Prediction in Display Advertising %A Ling Yan %A Wu-Jun Li %A Gui-Rong Xue %A Dingyi Han %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-yan14 %I PMLR %P 802--810 %U https://proceedings.mlr.press/v32/yan14.html %V 32 %N 2 %X In display advertising, click through rate(CTR) prediction is the problem of estimating the probability that an advertisement (ad) is clicked when displayed to a user in a specific context. Due to its easy implementation and promising performance, logistic regression(LR) model has been widely used for CTR prediction, especially in industrial systems. However, it is not easy for LR to capture the nonlinear information, such as the conjunction information, from user features and ad features. In this paper, we propose a novel model, called coupled group lasso(CGL), for CTR prediction in display advertising. CGL can seamlessly integrate the conjunction information from user features and ad features for modeling. Furthermore, CGL can automatically eliminate useless features for both users and ads, which may facilitate fast online prediction. Scalability of CGL is ensured through feature hashing and distributed implementation. Experimental results on real-world data sets show that our CGL model can achieve state-of-the-art performance on web-scale CTR prediction tasks.
RIS
TY - CPAPER TI - Coupled Group Lasso for Web-Scale CTR Prediction in Display Advertising AU - Ling Yan AU - Wu-Jun Li AU - Gui-Rong Xue AU - Dingyi Han BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/06/18 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-yan14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 2 SP - 802 EP - 810 L1 - http://proceedings.mlr.press/v32/yan14.pdf UR - https://proceedings.mlr.press/v32/yan14.html AB - In display advertising, click through rate(CTR) prediction is the problem of estimating the probability that an advertisement (ad) is clicked when displayed to a user in a specific context. Due to its easy implementation and promising performance, logistic regression(LR) model has been widely used for CTR prediction, especially in industrial systems. However, it is not easy for LR to capture the nonlinear information, such as the conjunction information, from user features and ad features. In this paper, we propose a novel model, called coupled group lasso(CGL), for CTR prediction in display advertising. CGL can seamlessly integrate the conjunction information from user features and ad features for modeling. Furthermore, CGL can automatically eliminate useless features for both users and ads, which may facilitate fast online prediction. Scalability of CGL is ensured through feature hashing and distributed implementation. Experimental results on real-world data sets show that our CGL model can achieve state-of-the-art performance on web-scale CTR prediction tasks. ER -
APA
Yan, L., Li, W., Xue, G. & Han, D.. (2014). Coupled Group Lasso for Web-Scale CTR Prediction in Display Advertising. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(2):802-810 Available from https://proceedings.mlr.press/v32/yan14.html.

Related Material