FairDrop: Biased Edge Dropout for Enhancing Fairness in Graph Representation Learning
- 10 December 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 3 (3) , 344-354
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 random walk model for producing node embeddings, and to a graph convolutional network for link prediction. We prove that the proposed algorithm can successfully improve the fairness of all models with only a small or negligible drop in accuracy, and compares favourably with existing state-of-the-art solutions. In an ablation study, we demonstrate that our algorithm can flexibly interpolate between biasing towards fairness and an unbiased edge dropout. Furthermore, to better evaluate the gains, we propose a new dyadic group definition to measure the bias of a link prediction task when paired with group-based fairness metrics. In particular, we extend the metric used to measure the bias in node embeddings by taking account of the graph structure.Keywords
All Related Versions
Funding Information
- Engineering and Physical Sciences Research Council (EP/M026981/1, EP/T021063/1, EP/T024917/1)
This publication has 33 references indexed in Scilit:
- Certifying and Removing Disparate Impact2015
- Automated Experiments on Ad Privacy SettingsProceedings on Privacy Enhancing Technologies, 2015
- DeepWalk2014
- A multidisciplinary survey on discrimination analysisThe Knowledge Engineering Review, 2013
- Fairness through awareness2012
- The Spread of Behavior in an Online Social Network ExperimentScience, 2010
- ArnetMiner2008
- The link‐prediction problem for social networksJournal of the American Society for Information Science and Technology, 2007
- Birds of a Feather: Homophily in Social NetworksAnnual Review of Sociology, 2001
- Randomized Response: A Survey Technique for Eliminating Evasive Answer BiasJournal of the American Statistical Association, 1965