Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GNNUERS: Fairness Explanation in GNNs for Recommendation via Counterfactual Reasoning

Giacomo Medda, Francesco Fabbri, Mirko Marras, Ludovico Boratto, Gianni Fenu

TL;DR

A novel algorithm that leverages counterfactuality methods to discover user unfairness explanations in the form of user-item interactions in the form of bipartite graph interactions is devised.

Abstract

Nowadays, research into personalization has been focusing on explainability and fairness. Several approaches proposed in recent works are able to explain individual recommendations in a post-hoc manner or by explanation paths. However, explainability techniques applied to unfairness in recommendation have been limited to finding user/item features mostly related to biased recommendations. In this paper, we devised a novel algorithm that leverages counterfactuality methods to discover user unfairness explanations in the form of user-item interactions. In our counterfactual framework, interactions are represented as edges in a bipartite graph, with users and items as nodes. Our bipartite graph explainer perturbs the topological structure to find an altered version that minimizes the disparity in utility between the protected and unprotected demographic groups. Experiments on four real-world graphs coming from various domains showed that our method can systematically explain user unfairness on three state-of-the-art GNN-based recommendation models. Moreover, an empirical evaluation of the perturbed network uncovered relevant patterns that justify the nature of the unfairness discovered by the generated explanations. The source code and the preprocessed data sets are available at https://github.com/jackmedda/RS-BGExplainer.

GNNUERS: Fairness Explanation in GNNs for Recommendation via Counterfactual Reasoning

TL;DR

A novel algorithm that leverages counterfactuality methods to discover user unfairness explanations in the form of user-item interactions in the form of bipartite graph interactions is devised.

Abstract

Nowadays, research into personalization has been focusing on explainability and fairness. Several approaches proposed in recent works are able to explain individual recommendations in a post-hoc manner or by explanation paths. However, explainability techniques applied to unfairness in recommendation have been limited to finding user/item features mostly related to biased recommendations. In this paper, we devised a novel algorithm that leverages counterfactuality methods to discover user unfairness explanations in the form of user-item interactions. In our counterfactual framework, interactions are represented as edges in a bipartite graph, with users and items as nodes. Our bipartite graph explainer perturbs the topological structure to find an altered version that minimizes the disparity in utility between the protected and unprotected demographic groups. Experiments on four real-world graphs coming from various domains showed that our method can systematically explain user unfairness on three state-of-the-art GNN-based recommendation models. Moreover, an empirical evaluation of the perturbed network uncovered relevant patterns that justify the nature of the unfairness discovered by the generated explanations. The source code and the preprocessed data sets are available at https://github.com/jackmedda/RS-BGExplainer.
Paper Structure (25 sections, 7 equations, 4 figures, 3 tables)

This paper contains 25 sections, 7 equations, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Beyond Predictions Explanation in Recommendation and on Graphs
  4. Explainability and User Fairness in Recommendation
  5. Explainability in GNNs
  6. Problem Formulation
  7. Recommendation Task
  8. Unfairness Explanation Task
  9. Methodology: GNNUERS
  10. Bipartite Graph Perturbation
  11. Perturbed Graph Generation
  12. Loss Function Optimization
  13. Gradient Deactivation
  14. Resources Usage
  15. Evaluation
  16. ...and 10 more sections

Figures (4)

  • Figure 1: Distribution of $\Delta$NDCG between younger and older users subgroups, randomly sampled 100 times. A Wilcoxon signed-rank test is performed between each pair of boxes and the respective p-value is shown if it is lower than $\frac{0.05}{m}$ according to the Bonferroni correction, where $m$ is the number of pairwise comparisons performed for each model.
  • Figure 2: Distribution of $\Delta$NDCG between males and females users subgroups, randomly sampled 100 times. A Wilcoxon signed-rank test is performed between each pair of boxes and the respective p-value is shown if it is lower than $\frac{0.05}{m}$ according to the Bonferroni correction, where $m$ is the number of pairwise comparisons performed for each model.
  • Figure 3: Deleted edges distribution (Del Edges Distribution) over the quartiles (Q1-Q2-Q3-Q4) defined for each age group through sorting the nodes by each graph property. The edges were deleted applying GNNUERS on NGCF.
  • Figure 4: Deleted edges distribution (Del Edges Distribution) over the quartiles (Q1-Q2-Q3-Q4) defined for each gender group through sorting the nodes by each graph property. The edges were deleted applying GNNUERS on NGCF.

Theorems & Definitions (2)

  • Definition 1: Counterfactual Explanation of Recommendation Unfairness in GNNs
  • Definition 2: Optimal Counterfactual Explanation of Recommendation Unfairness in GNNs