Optimization and Analysis of the pAp@k Metric for Recommender Systems

Gaurush Hiranandani, Warut Vijitbenjaronk, Sanmi Koyejo, Prateek Jain
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:4260-4270, 2020.

Abstract

Modern recommendation and notification systems must be robust to data imbalance, limitations on the number of recommendations/notifications, and heterogeneous engagement profiles across users. The pAp@k metric, which combines the partial-AUC and the precision@k metrics, was recently proposed to evaluate such recommendation systems and has been used in real-world deployments. Conceptually, pAp@k measures the probability of correctly ranking a top-ranked positive instance over top-ranked negative instances. Due to the combinatorial aspect surfaced by top-ranked points, little is known about the characteristics and optimization methods of pAp@k. In this paper, we analyze the learning-theoretic properties of pAp@k, particularly its benefits in evaluating modern recommender systems, and propose novel surrogates that are consistent under certain data regularity conditions. We then provide gradient descent based algorithms to optimize the surrogates directly. Our analysis and experimental evaluation suggest that pAp@k indeed exhibits a certain dual behavior with respect to partial-AUC and precision@k. Moreover, the proposed methods outperform all the baselines in various applications. Taken together, our results motivate the use of pAp@k for large-scale recommender systems with heterogeneous user-engagement.

Cite this Paper


BibTeX
@InProceedings{pmlr-v119-hiranandani20a, title = {Optimization and Analysis of the p{A}p@k Metric for Recommender Systems}, author = {Hiranandani, Gaurush and Vijitbenjaronk, Warut and Koyejo, Sanmi and Jain, Prateek}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {4260--4270}, year = {2020}, editor = {III, Hal Daumé and Singh, Aarti}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/hiranandani20a/hiranandani20a.pdf}, url = {http://proceedings.mlr.press/v119/hiranandani20a.html}, abstract = {Modern recommendation and notification systems must be robust to data imbalance, limitations on the number of recommendations/notifications, and heterogeneous engagement profiles across users. The pAp@k metric, which combines the partial-AUC and the precision@k metrics, was recently proposed to evaluate such recommendation systems and has been used in real-world deployments. Conceptually, pAp@k measures the probability of correctly ranking a top-ranked positive instance over top-ranked negative instances. Due to the combinatorial aspect surfaced by top-ranked points, little is known about the characteristics and optimization methods of pAp@k. In this paper, we analyze the learning-theoretic properties of pAp@k, particularly its benefits in evaluating modern recommender systems, and propose novel surrogates that are consistent under certain data regularity conditions. We then provide gradient descent based algorithms to optimize the surrogates directly. Our analysis and experimental evaluation suggest that pAp@k indeed exhibits a certain dual behavior with respect to partial-AUC and precision@k. Moreover, the proposed methods outperform all the baselines in various applications. Taken together, our results motivate the use of pAp@k for large-scale recommender systems with heterogeneous user-engagement.} }
Endnote
%0 Conference Paper %T Optimization and Analysis of the pAp@k Metric for Recommender Systems %A Gaurush Hiranandani %A Warut Vijitbenjaronk %A Sanmi Koyejo %A Prateek Jain %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-hiranandani20a %I PMLR %P 4260--4270 %U http://proceedings.mlr.press/v119/hiranandani20a.html %V 119 %X Modern recommendation and notification systems must be robust to data imbalance, limitations on the number of recommendations/notifications, and heterogeneous engagement profiles across users. The pAp@k metric, which combines the partial-AUC and the precision@k metrics, was recently proposed to evaluate such recommendation systems and has been used in real-world deployments. Conceptually, pAp@k measures the probability of correctly ranking a top-ranked positive instance over top-ranked negative instances. Due to the combinatorial aspect surfaced by top-ranked points, little is known about the characteristics and optimization methods of pAp@k. In this paper, we analyze the learning-theoretic properties of pAp@k, particularly its benefits in evaluating modern recommender systems, and propose novel surrogates that are consistent under certain data regularity conditions. We then provide gradient descent based algorithms to optimize the surrogates directly. Our analysis and experimental evaluation suggest that pAp@k indeed exhibits a certain dual behavior with respect to partial-AUC and precision@k. Moreover, the proposed methods outperform all the baselines in various applications. Taken together, our results motivate the use of pAp@k for large-scale recommender systems with heterogeneous user-engagement.
APA
Hiranandani, G., Vijitbenjaronk, W., Koyejo, S. & Jain, P.. (2020). Optimization and Analysis of the pAp@k Metric for Recommender Systems. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:4260-4270 Available from http://proceedings.mlr.press/v119/hiranandani20a.html.

Related Material