Dirichlet Energy Constrained Learning for Deep Graph Neural Networks
Authors: Kaixiong Zhou, Xiao Huang, Daochen Zha, Rui Chen, Li Li, Soo-Hyun Choi, Xia Hu
NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we empirically evaluate the effectiveness of EGNN on real-world datasets. We aim to answer the following questions. Q1: How does our EGNN compare with the state-of-the-art deep GNN models? Q2: Whether or not the Dirichlet energy at each layer of EGNN satisfies the constrained learning? Q3: How does each component of EGNN affect the model performance? Q4: How do the model hyperparameters impact the performance of EGNN? |
| Researcher Affiliation | Collaboration | Kaixiong Zhou Rice University Kaixiong.Zhou@rice.edu Xiao Huang The Hong Kong Polytechnic University xiaohuang@comp.polyu.edu.hk Daochen Zha Rice University Daochen.Zha@rice.edu Rui Chen Samsung Research America rui.chen1@samsung.com Li Li Samsung Research America li.li1@samsung.com Soo-Hyun Choi Samsung Electronics soohyunc@gmail.com Xia Hu Rice University xia.hu@rice.edu |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper states 'We implement all the baselines using Pytorch Geometric [55] based on their official implementations,' but does not explicitly state that the source code for their own method (EGNN) is publicly available or provide a link. |
| Open Datasets | Yes | Following the practice of previous work, we evaluate EGNN by performing node classification on four benchmark datasets: Cora, Pubmed [52], Coauthor-Physics [53] and Ogbn-arxiv [54]. |
| Dataset Splits | Yes | We choose hyperparameters cmax, cmin, γ and b based on the validation set. |
| Hardware Specification | No | The paper does not provide any specific hardware details (like exact GPU/CPU models or types) used for running its experiments. |
| Software Dependencies | No | The paper mentions 'Pytorch Geometric [55]' but does not specify a version number for this or any other software dependency. |
| Experiment Setup | Yes | We choose hyperparameters cmax, cmin, γ and b based on the validation set. For the weight initialization, we set cmax to be 1 for all the datasets; that is, the trainable weights are initialized as identity matrices at all the graph convolutional layers. The loss hyperparameter γ is 20 in Cora, Pubmed and Coauthor-Physics to strictly regularize towards the orthogonal matrix; and it is 10 4 in Ogbn-arxiv to improve the model s learning ability. For the lower-bounded residual connection, we choose residual strength cmin from range [0.1, 0.75] and list the details in Appendix. The trainable shift b is initialized with 10 in Cora and Pubmed; it is initialized to 5 and 1 in Coauthor-Physics and Ogbn-arxiv, respectively. |