Please use this identifier to cite or link to this item:
Title: Adversarial Personalized Ranking for Recommendation
Authors: Xiangnan He 
Zhankui He
Xiaoyu Du
Tat-Seng Chua 
Keywords: Adversarial training
Item recommendation
Matrix factorization
Pairwise learning
Personalized ranking
Issue Date: 12-Jul-2018
Publisher: Association for Computing Machinery, Inc
Citation: Xiangnan He, Zhankui He, Xiaoyu Du, Tat-Seng Chua (2018-07-12). Adversarial Personalized Ranking for Recommendation. ACM SIGIR Conference on Information Retrieval 2018 : 355-364. ScholarBank@NUS Repository.
Abstract: Item recommendation is a personalized ranking task. To this end, many recommender systems optimize models with pairwise ranking objectives, such as the Bayesian Personalized Ranking (BPR). Using matrix Factorization (MF) - the most widely used model in recommendation - as a demonstration, we show that optimizing it with BPR leads to a recommender model that is not robust. In particular, we find that the resultant model is highly vulnerable to adversarial perturbations on its model parameters, which implies the possibly large error in generalization. To enhance the robustness of a recommender model and thus improve its generalization performance, we propose a new optimization framework, namely Adversarial Personalized Ranking (APR). In short, our APR enhances the pairwise ranking method BPR by performing adversarial training. It can be interpreted as playing a minimax game, where the minimization of the BPR objective function meanwhile defends an adversary, which adds adversarial perturbations on model parameters to maximize the BPR objective function. To illustrate how it works, we implement APR on MF by adding adversarial perturbations on the embedding vectors of users and items. Extensive experiments on three public real-world datasets demonstrate the effectiveness of APR - by optimizing MF with APR, it outperforms BPR with a relative improvement of 11.2% on average and achieves state-of-the-art performance for item recommendation. Our implementation is available at: \url © 2018 ACM.
Source Title: ACM SIGIR Conference on Information Retrieval 2018
ISBN: 9781450356572
DOI: 10.1145/3209978.3209981
Appears in Collections:Elements
Staff Publications

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Adversarial Personalized Ranking for Recommendation.pdf185.59 kBAdobe PDF




checked on Dec 9, 2022

Page view(s)

checked on Dec 1, 2022


checked on Dec 1, 2022

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.