Understanding Oversquashing in GNNs through the Lens of Effective Resistance

Authors: Mitchell Black, Zhengchao Wan, Amir Nayyeri, Yusu Wang

ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In Section 5, we empirically demonstrate that our rewiring technique is effective in alleviating oversquashing. and 5. Experiments
Researcher Affiliation Academia 1School of Electrical Engineering and Computer Science, Oregon State University, Corvallis, Oregon, USA 2Halıcıoˇglu Data Science Institute, University of California San Diego, San Diego, California, USA.
Pseudocode No The paper describes the 'Greedy Total Resistance (GTR) rewiring' heuristic in paragraph form within Section 4 but does not provide a structured pseudocode or algorithm block.
Open Source Code Yes A Py Torch Geometric implementation of the GTR algorithm is available online2. and footnote 2: https://github.com/blackmit/gtr_rewiring
Open Datasets Yes All datasets are from the TUDataset (Morris et al., 2020).
Dataset Splits Yes We use randomly generated 80%/10%/10% train/validation/test splits of the data.
Hardware Specification No The paper does not provide specific hardware details (such as exact GPU/CPU models, processor types, or memory amounts) used for running its experiments. It does not mention any hardware specifications.
Software Dependencies No The paper mentions 'Py Torch Geometric' and 'Torch' but does not provide specific version numbers for these software dependencies or any other ancillary software.
Experiment Setup Yes Table 4: Hyperparameters for Graph Classifcation. These are consistent across all GNN types. These are the same as used in the experiments in (Karhadkar et2022) Hyperparameters Number of Hidden Layers 4 Dimension of Hidden Layers 64 Dropout 0.5 Learning Rate 1.0 10 3