A Deep Generative Recommendation Method for Unbiased Learning from Implicit Feedback

被引:2
|
作者
Gupta, Shashank [1 ]
Oosterhuis, Harrie [2 ]
de Rijke, Maarten [1 ]
机构
[1] Univ Amsterdam, Amsterdam, Netherlands
[2] Radboud Univ Nijmegen, Nijmegen, Netherlands
关键词
Variational autoencoder; Implicit feedback; Selection bias;
D O I
10.1145/3578337.3605114
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Variational autoencoders (VAEs) are the state-of-the-art model for recommendation with implicit feedback signals. Unfortunately, implicit feedback suffers from selection bias, e.g., popularity bias, position bias, etc., and as a result, training from such signals produces biased recommendation models. Existing methods for debiasing the learning process have not been applied in a generative setting. We address this gap by introducing an inverse propensity scoring (IPS) based method for training VAEs from implicit feedback data in an unbiased way. Our IPS-based estimator for the VAE training objective, VAE-IPS, is provably unbiased w.r.t. selection bias. Our experimental results show that the proposed VAE-IPS model reaches significantly higher performance than existing baselines. Our contributions enable practitioners to combine state-of-the-art VAE recommendation techniques with the advantages of bias mitigation for implicit feedback.
引用
收藏
页码:87 / 93
页数:7
相关论文
共 50 条
  • [41] Reinforcing Recommendation Using Implicit Negative Feedback
    Lee, Danielle H.
    Brusilovsky, Peter
    USER MODELING, ADAPTATION, AND PERSONALIZATION, PROCEEDINGS, 2009, 5535 : 422 - 427
  • [42] Modeling Dynamic Missingness of Implicit Feedback for Recommendation
    Wang, Menghan
    Gong, Mingming
    Zheng, Xiaolin
    Zhang, Kun
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), 2018, 31
  • [43] Contextual Restaurant Recommendation Utilizing Implicit Feedback
    Kuo, Wei-Ti
    Wang, Yu-Chun
    Tsai, Richard Tzong-Han
    Hsu, Jane Yung-jen
    2015 24TH WIRELESS AND OPTICAL COMMUNICATION CONFERENCE (WOCC), 2015, : 170 - 174
  • [44] A Boosting Algorithm for Item Recommendation with Implicit Feedback
    Liu, Yong
    Zhao, Peilin
    Sun, Aixin
    Miao, Chunyan
    PROCEEDINGS OF THE TWENTY-FOURTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI), 2015, : 1792 - 1798
  • [45] Personalized Recommendation Considering Secondary Implicit Feedback
    Liu, Siyuan
    Wu, Qiong
    Miao, Chunyan
    2018 IEEE INTERNATIONAL CONFERENCE ON AGENTS (ICA), 2018, : 87 - 92
  • [46] Correlated Matrix Factorization for Recommendation with Implicit Feedback
    He, Yuan
    Wang, Cheng
    Jiang, Changjun
    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2019, 31 (03) : 451 - 464
  • [47] Matrix factorization for recommendation with explicit and implicit feedback
    Chen, Shulong
    Peng, Yuxing
    KNOWLEDGE-BASED SYSTEMS, 2018, 158 : 109 - 117
  • [48] Personalized Game Recommendation Based on Implicit Feedback
    Yu D.-J.
    Chen C.
    Wu J.-H.
    Chen Y.-W.
    Tien Tzu Hsueh Pao/Acta Electronica Sinica, 2018, 46 (11): : 2626 - 2632
  • [49] Unbiased Generative Semi-Supervised Learning
    Fox-Roberts, Patrick
    Rosten, Edward
    JOURNAL OF MACHINE LEARNING RESEARCH, 2014, 15 : 367 - 443
  • [50] Unbiased generative semi-supervised learning
    1600, Microtome Publishing (15):