Upper bounds on misclassification rate in DP-GCNs are derived as a function of subsampling probability p_s, with feasible ranges of p_s identified to balance privacy and utility.
Training differen- tially private graph neural networks with random walk sampling
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Misclassification Rate and Privacy-Utility Trade-offs in Graph Convolutional Networks via Subsampling Stability
Upper bounds on misclassification rate in DP-GCNs are derived as a function of subsampling probability p_s, with feasible ranges of p_s identified to balance privacy and utility.