Implicit Graph Neural Networks

Authors: Fangda Gu, Heng Chang, Wenwu Zhu, Somayeh Sojoudi, Laurent El Ghaoui

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on a comprehensive range of tasks show that IGNNs consistently capture long-range dependencies and outperform the state-of-the-art GNN models.
Researcher Affiliation Academia 1Department of Electrical Engineering and Computer Sciences, University of California at Berkeley 2Tsinghua-Berkeley Shenzhen Institute, Tsinghua University 3Department of Computer Science and Technology, Tsinghua University
Pseudocode No The paper includes mathematical equations and descriptions of the model, but it does not contain a clearly labeled 'Pseudocode' or 'Algorithm' block with structured steps.
Open Source Code Yes Code available at https://github.com/Swiftie H/IGNN.
Open Datasets Yes To further manifest the scalability of IGNN towards larger graphs, we conduct experiments on a large multi-label node classification data set, namely the Amazon product co-purchasing network data set (Yang and Leskovec, 2015) 3. The data set renders products as nodes and co-purchases as edges but provides no input features. ... 3http://snap.stanford.edu/data/#amazon
Dataset Splits Yes We use a small training set, validation set, and test set with only 20, 100, and 200 nodes, respectively. ... The train/valid/test split is consistent with Graph Sage (Hamilton et al., 2017). ... 10-fold cross-validation with LIB-SVM (Chang and Lin, 2011) is conducted.
Hardware Specification No The paper does not provide specific details about the hardware used for running the experiments (e.g., CPU models, GPU models, memory specifications).
Software Dependencies No The paper mentions 'LIB-SVM (Chang and Lin, 2011)' and alludes to 'autograd software', but it does not specify version numbers for any software dependencies.
Experiment Setup Yes Detailed description of the data sets, our preprocessing procedure, hyper-parameters, and other information of experiments can be found in Appendix E.