Corrective unlearning for graph neural networks is achieved by alternating contrastive separation of affected neighborhoods with asymmetric gradient ascent and descent, using as little as 5 percent of the manipulated set.
Certifiable robustness to graph perturbations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Cognac Shot To Forget Bad Memories: Corrective Unlearning for Graph Neural Networks
Corrective unlearning for graph neural networks is achieved by alternating contrastive separation of affected neighborhoods with asymmetric gradient ascent and descent, using as little as 5 percent of the manipulated set.