D2Match: Leveraging Deep Learning and Degeneracy for Subgraph Matching
Authors: Xuanzhou Liu, Lin Zhang, Jiaqi Sun, Yujiu Yang, Haiqin Yang
ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally, we conduct extensive experiments to show the superior performance of our D2Match and confirm that our D2Match indeed exploits the subtrees and differs from existing GNNs-based subgraph matching methods that depend on memorizing the data distribution divergence. |
| Researcher Affiliation | Collaboration | 1Shenzhen International Graduate School, Tsinghua University, Shenzhen, China 2International Digital Economy Academy (IDEA). Work done when Xuanzhou was interned at IDEA. Correspondence to: Yujiu Yang <yang.yujiu@sz.tsinghua.edu.cn>, Haiqin Yang <hqyang@ieee.org>. |
| Pseudocode | Yes | The pseudo-code of D2Match is outlined as follows: Algorithm 1 The D2Match algorithm |
| Open Source Code | No | The python implementation of D2Match will be available at https://github.com/Xuanzhou Liu/D2Match-ICML23 |
| Open Datasets | Yes | We first generate synthetic data by utilizing ER-random graphs and WS-random graphs (Rex et al., 2020). [...] For the real-world data, we follow the setting in (Rex et al., 2020), including Cox2, Enzymes, Proteins, IMDB-Binary, MUTAG, Aids, and First MMDB. |
| Dataset Splits | Yes | We split each dataset into training and testing at a ratio of 4 : 1 and report the average classification accuracy under the five-fold cross-validation. |
| Hardware Specification | No | The paper does not provide specific details on the hardware used for running experiments, such as CPU or GPU models, or memory specifications. |
| Software Dependencies | No | The paper mentions using Adam as an optimizer but does not specify software dependencies like programming languages, libraries, or frameworks with version numbers (e.g., Python 3.x, PyTorch 1.x). |
| Experiment Setup | Yes | We set all models with adjustable number of layers to 5 layers, and set the hidden dimension to 10 to avoid overfitting. [...] Both our model and all baselines use the Adam as optimizer and set the learning rate to 3e-4. [...] We test the effect of the depth of a subtree, i.e., the number of the hidden layers, and change it from 1 to 7. [...] We vary K from 1 to 7 and show the results in Fig. 2(b). |