FairDrop: Biased Edge Dropout for Enhancing Fairness in Graph Representation Learning

Spinelli, Indro; Scardapane, Simone; Hussain, Amir; Uncini, Aurelio

Abstract

Graph representation learning has become a ubiquitous component in many scenarios, ranging from social network analysis to energy forecasting in smart grids. In several applications, ensuring the fairness of the node (or graph) representations with respect to some protected attributes is crucial for their correct deployment. Yet, fairness in graph deep learning remains underexplored, with few solutions available. In particular, the tendency of similar nodes to cluster on several real-world graphs (i.e., homophily) can dramatically worsen the fairness of these procedures. In this article, we propose a novel biased edge dropout algorithm (FairDrop) to counter-act homophily and improve fairness in graph representation learning. FairDrop can be plugged in easily on many existing algorithms, is efficient, adaptable, and can be combined with other fairness-inducing solutions. After describing the general algorithm, we demonstrate its application on two benchmark tasks, specifically, as a ran...

Citation

Spinelli, I., Scardapane, S., Hussain, A., & Uncini, A. (2022). FairDrop: Biased Edge Dropout for Enhancing Fairness in Graph Representation Learning. IEEE Transactions on Artificial Intelligence, 3(3), 344-354. https://doi.org/10.1109/tai.2021.3133818

• Journal Article Type: Article
• Online Publication Date: Dec 10, 2021
• Publication Date: 2022-06
• Deposit Date: Nov 21, 2022
• Journal: IEEE Transactions on Artificial Intelligence
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Volume: 3
• Issue: 3
• Pages: 344-354
• DOI: https://doi.org/10.1109/tai.2021.3133818
• Keywords: Fairness, graph embedding, graph neural network, graph representation learning, link prediction
• Public URL: http://researchrepository.napier.ac.uk/Output/2963037