MGNNI: Multiscale Graph Neural Networks with Implicit Layers

Authors: Juncheng Liu, Bryan Hooi, Kenji Kawaguchi, Xiaokui Xiao

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct comprehensive experiments for both node classification and graph classification to show that MGNNI outperforms representative baselines and has a better ability for multiscale modeling and capturing of long-range dependencies.
Researcher Affiliation Academia Juncheng Liu Bryan Hooi Kenji Kawaguchi Xiaokui Xiao National University of Singapore {juncheng,bhooi,kenji,xiaoxk}@comp.nus.edu.sg
Pseudocode No The paper describes algorithms using mathematical equations and textual explanations but does not include structured pseudocode or algorithm blocks.
Open Source Code Yes The implementation can be found at https://github.com/liu-jc/MGNNI
Open Datasets Yes We use the standard train/test/val splits as in Pei et al. [23]. See the detailed setting in Appendix C.2. ... The same train/val/test split are used as in Hamilton et al. [14].
Dataset Splits Yes We use the standard train/test/val splits as in Pei et al. [23]. ... 10-fold Cross-validation is conducted as [31]
Hardware Specification No The main body of the paper does not specify the hardware used (e.g., GPU models, CPU types, or cloud instances). While Appendix C is mentioned for resource details in the checklist, it is not provided in the given text.
Software Dependencies No The paper mentions 'Py Torch [22]' without a specific version number and does not list other software dependencies with their versions.
Experiment Setup No The main paper refers to Appendix C for detailed experimental settings ('See the detailed setting in Appendix C.2.'), indicating that specific hyperparameters and training configurations are not included in the provided text.