Debiasing Neighbor Aggregation for Graph Neural Network in Recommender Systems

TL;DR

This paper introduces Navip, a method to reduce exposure bias in graph neural networks for recommender systems by applying inverse propensity weighting during neighbor aggregation, leading to significant performance improvements.

Contribution

It proposes a novel neighbor aggregation debiasing technique using inverse propensity scores in GNNs for recommender systems, addressing exposure bias during aggregation.

Findings

Performance improved up to 14.2% on datasets.

Effective reduction of exposure bias in GNN neighbor aggregation.

Validated on multiple public datasets.

Abstract

Graph neural networks (GNNs) have achieved remarkable success in recommender systems by representing users and items based on their historical interactions. However, little attention was paid to GNN's vulnerability to exposure bias: users are exposed to a limited number of items so that a system only learns a biased view of user preference to result in suboptimal recommendation quality. Although inverse propensity weighting is known to recognize and alleviate exposure bias, it usually works on the final objective with the model outputs, whereas GNN can also be biased during neighbor aggregation. In this paper, we propose a simple but effective approach, neighbor aggregation via inverse propensity (Navip) for GNNs. Specifically, given a user-item bipartite graph, we first derive propensity score of each user-item interaction in the graph. Then, inverse of the propensity score with Laplacian normalization is applied to debias neighbor aggregation from exposure bias. We validate the effectiveness of our approach through our extensive experiments on two public and Amazon Alexa datasets where the performance enhances up to 14.2%.