Towards More Practical Adversarial Attacks on Graph Neural Networks

Authors: Jiaqi Ma, Shuangrui Ding, Qiaozhu Mei

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that the proposed procedure can significantly increase the mis-classification rate of common GNNs on real-world data without access to model parameters nor predictions.
Researcher Affiliation Academia School of Information, University of Michigan, Ann Arbor, Michigan, USA. Department of EECS, University of Michigan, Ann Arbor, Michigan, USA.
Pseudocode Yes Algorithm 1: The GC-RWCS Strategy for Node Selection.
Open Source Code No The paper does not provide an explicit statement about releasing source code or a link to a code repository for the methodology described.
Open Datasets Yes We adopt three citation networks, Citeseer, Cora, and Pubmed, which are standard node classification benchmark datasets [26].
Dataset Splits Yes Following the setup of JK-Net [25], we randomly split each dataset by 60%, 20%, and 20% for training, validation, and testing.
Hardware Specification No The paper does not provide any specific details about the hardware used to run the experiments, such as GPU models, CPU types, or memory specifications.
Software Dependencies No The paper mentions using GCN and JK-Net models and following hyper-parameter setups from a prior work [25], but it does not specify any software dependencies with version numbers, such as programming languages, libraries, or frameworks (e.g., Python version, PyTorch/TensorFlow version).
Experiment Setup Yes We set the number of layers for GCN as 2 and the number of layers for both JK-Concat and JK-Maxpool as 7. The hidden size of each layer is 32. For the proposed GC-RWCS strategy, we fix the number of step L = 4, the neighbor-hop parameter k = 1 and the parameter l = 30 for the binarized f M in Eq. (4) for all models on all datasets.