CONC: Complex-noise-resistant Open-set Node Classification with Adaptive Noise Detection

Authors: Qin Zhang, Jiexin Lu, Xiaowei Li, Huisi Wu, Shirui Pan, Junyang Chen

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

Reproducibility Variable Result LLM Response
Research Type Experimental The experimental results demonstrate the superiority of our method. We designed experiments to evaluate CONC, focusing primarily on: comparison of open-set classification, robustness analysis, and ablation study.
Researcher Affiliation Academia 1College of Computer Science and Software Engineering, Shenzhen University, China 2School of Information and Communication Technology, Griffith University, Australia
Pseudocode No The paper does not contain pseudocode or clearly labeled algorithm blocks.
Open Source Code No The relevant code will be made publicly accessible online.
Open Datasets Yes To evaluate the performance of the proposed framework for robust open-set node classification, we performed experiments on four widely-used citation network benchmark datasets: Cora, Citeseer, Coauthor-CS, and Ogbn-arxiv. The datasets statistics are presented in the Appendix.
Dataset Splits Yes Among the known classes, 70% of nodes were allocated for training, 10% for validation, and the remaining 20% for testing.
Hardware Specification No The paper does not explicitly describe the hardware used to run its experiments.
Software Dependencies No The CONC framework is implemented using Py Torch, and network optimization leverages adaptive moment estimation with a learning rate of 10-3.
Experiment Setup Yes For CONC, we use GCN [Kipf and Welling, 2016] as the backbone network for the encoder ϕb, with four hidden layers of 128 dimensions. The feature decoder ψf and the degree decoder ψg are configured with three-layer and four-layer multilayer perceptron with a dimension of 128. To train the open-set node classifier ψc, an additional linear layer is added for it. We employed a grid search to determine the optimal values for thresholds τ, τ1, τ2, exploring the range from 0 to 1 with a step size of 10-1. The CONC framework is implemented using Py Torch, and network optimization leverages adaptive moment estimation with a learning rate of 10-3. Balance parameters γ1 and γ2 were also identified through a grid search, considering values from 10-3 to 101 with a step size of 101.