arrow
Return

Learning instrumental variable representation for debiasing in recommender systems

delete2025-08-12
delete0
PRE
AI
H
Huang Zhi-rong
S
Shichao Zhang
D
Debo Cheng
J
Jiuyong Li
刘琳 (Lin Liu)
G
Guangquan Lu
G
Guixian Zhang
DOI:10.1016/j.neunet.2025.107977delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• DIVRS: A debiasing method learning SIV representation to ease IV selection. • DIVRS-GCN: An extension of DIVRS for GCN, integrating OPR for effective adaptation. • Experiments show DIVRS and DIVRS-GCN mitigate bias and improve recommendation quality.
Keywords:
Recommender systems
Instrumental variable
Confounding bias
Latent confounders

Journal

Neural Networks cover
Neural Networks
IF:
6.3
Papers:
7.8K
Citations:
3.0W

Organization

U
unisa stem
Scholars:
12
Papers: 10
Citations: 0
S
School of Computer Science and Technology
Scholars:
1.5K
Papers: 566
Citations: 0
researcher View more organizations