Poincaré Differential Privacy for Hierarchy-Aware Graph Embedding

Authors: Yuecen Wei, Haonan Yuan, Xingcheng Fu, Qingyun Sun, Hao Peng, Xianxian Li, Chunming Hu

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiment results on five real-world datasets demonstrate the proposed Poin DP s advantages of effective privacy protection while maintaining good performance on the node classification task.
Researcher Affiliation Academia Yuecen Wei1,2,3, Haonan Yuan1, Xingcheng Fu3*, Qingyun Sun1, Hao Peng1, Xianxian Li3, Chunming Hu1,2* 1Beijing Advanced Innovation Center for Big Data and Brain Computing, Beihang University, Beijing, China 2School of Software, Beihang University, Beijing, China 3Key Lab of Education Blockchain and Intelligent Technology, Ministry of Education, Guangxi Normal University, China
Pseudocode Yes Algorithm 1: Overall training process of Poin DP
Open Source Code Yes Code is available at https://github.com/WYLucency/Poin DP.
Open Datasets Yes For datasets (see Appendix B.1), we chose three citation networks (Cora, Citeseer and Pub Med) and two E-commerce networks in Amazon (Computers and Photo).
Dataset Splits Yes Our dataset split follows the Py Torch Geometric.
Hardware Specification Yes We conducted the experiments with NVIDIA Ge Force RTX 3090 with 16GB of Memory.
Software Dependencies No The paper mentions 'Py Torch Geometric' but does not provide specific version numbers for this or any other software dependencies.
Experiment Setup Yes The learning rate lr is 0.005, the privacy budget ϵ to be [0, 1], and the training iterations E to be 200.