Adversarial Contrastive Graph Masked AutoEncoder Against Graph Structure and Feature Dual Attacks

Authors: Weixuan Shen, Xiaobo Shen, Shirui Pan

AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on node classification and clustering tasks demonstrate the effectiveness of the proposed ACGMAE, especially under graph structure and feature dual attacks.
Researcher Affiliation Academia 1Nanjing University of Science and Technology, Nanjing, China 2Griffith University, Gold Coast, Australia
Pseudocode Yes Algorithm 1: Algorithm of ACGMAE
Open Source Code No The paper does not contain an explicit statement about releasing the source code for the methodology described, nor does it provide a link to a code repository.
Open Datasets Yes We evaluate ACGMAE and baselines on three benchmark datasets, i.e., Cora, Citeseer, Pubmed (Jin et al. 2020)
Dataset Splits Yes For node classification, we randomly select 10% nodes for training, 10% nodes for validation, and the remaining for testing.
Hardware Specification Yes The experiments are performed on a Ubuntu Enterprise 64Bit Linux workstation with 128G memory and a NVIDIA A6000 GPU server.
Software Dependencies No The paper mentions that a two-layer GCN is employed as the encoder, but it does not specify any software names with version numbers (e.g., Python, PyTorch, TensorFlow, or specific libraries).
Experiment Setup Yes In the proposed ACGMAE, the learning rate and weight decay are searched from {0.01, 0.001, 0.0001} and {0.0001, 0.0005, 0.0001, 0.00005} respectively. The perturbation ratio X is searched from {0.1, 0.3, 0.5, 0.7, 0.9}, and the number of nearest neighbors and the number of clusters are searched from {10, 15, 20, 25, 30}. The coefficients α, β, and γ are searched from {0.01, 0.1, 0.5, 1, 3, 5}.