APL: Adversarial Pairwise Learning for Recommender Systems

作者：

Highlights：

• An adversarial pairwise learning model named APL is proposed for recommender systems.

• A differentiable procedure is adopted to replace the discrete item sampling.

• Three pairwise loss functions are evaluated under multiple recommendation scenarios.

• APL considerably improves the stability and convergence of adversarial learning.

摘要

•An adversarial pairwise learning model named APL is proposed for recommender systems.•A differentiable procedure is adopted to replace the discrete item sampling.•Three pairwise loss functions are evaluated under multiple recommendation scenarios.•APL considerably improves the stability and convergence of adversarial learning.

论文关键词：Adversarial learning,Pairwise ranking,Matrix factorization,Recommender systems

论文评审过程：Received 22 June 2018, Revised 12 October 2018, Accepted 13 October 2018, Available online 15 October 2018, Version of Record 24 October 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.10.024