<ul><li>Graph unlearning is essential for protecting user privacy by removing user data influence from trained graph models.</li><li>Recent developments in graph unlearning methods have focused on maintaining model performance while deleting user information, but changes in prediction distribution across sensitive groups have been observed.</li><li>Study shows that graph unlearning introduces bias, and a fair graph unlearning method, FGU, is proposed to address this issue.</li><li>FGU ensures privacy by training shard models on partitioned subgraphs and fairness by employing a bi-level debiasing process, achieving superior fairness without compromising privacy and accuracy.</li></ul>

Enabling Group Fairness in Graph Unlearning via Bi-level Debiasing

Discover more