On Structural Explanation of Bias in Graph Neural Networks

Published in Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2022

Recommended citation: In this work, we try to identify (explain) which edges inform the bias in GNNs decision making and propose a heuristics-based method to remove edges for enhancing fairness of GNNs. /files/pdf/research/StructFair.pdf