Convergent Privacy Framework for Multi-layer GNNs through Contractive Message Passing
Year: 2025
Type: conference-paper
Abstract: CARIBOU is a privacy-preserving GNN framework that couples a Contractive Graph Layer (CGL) with convergent privacy accounting, enabling deeper aggregation under edge-level DP (EDP) and node-level DP (NDP) with superior utility.
Authors: Zheng, Yu, Li, Chenang, Wang, Qingsong, Li, Zhou
Institution: University of California, Irvine
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7104274689
FWCI:
Cites:
Cited by:
Related to: