A Graph Kernel Based on the Jensen-Shannon Representation Alignment
Authors: Lu Bai, Zhihong Zhang, Chaoyan Wang, Xiao Bai, Edwin Hancock
IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments demonstrate that our kernel can easily outperform state-of-the-art graph kernels in terms of the classification accuracies. We demonstrate the performance of our new kernel on three standard graph datasets from computer vision databases. We report the average classification accuracies and standard errors for each kernel in Table.1. |
| Researcher Affiliation | Academia | Lu Bai1,2 , Zhihong Zhang3 , Chaoyan Wang4, Xiao Bai5, Edwin R. Hancock2 1School of Information, Central University of Finance and Economics, Beijing, China 2Department of Computer Science, University of York, York, UK 3Software School, Xiamen University, Xiamen, Fujian, China 4School of Contemporary Chinese Studies, University of Nottingham, Nottingham, UK 5School of Computer Science and Engineering, Beihang University, Beijing, China |
| Pseudocode | No | The paper does not contain any pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete access information (e.g., specific repository link, explicit code release statement) for the source code of the described methodology. |
| Open Datasets | Yes | BAR31, BSPHERE31 and GEOD31: The SHREC 3D Shape database consists of 15 classes and 20 individuals per class, that is 300 shapes [Biasotti et al., 2003]. |
| Dataset Splits | Yes | We perform 10-fold cross-validation using the C-Support Vector Machine (C-SVM) Classification to compute the classification accuracies, using LIBSVM [Chang and Lin, 2011]. We use nine samples for training and one for testing. |
| Hardware Specification | Yes | The CPU runtime is reported in Fig.2, as operated in Matlab R2011b on a 2.5GHz Intel 2-Core processor (i.e., i5-3210m). |
| Software Dependencies | Yes | The CPU runtime is reported in Fig.2, as operated in Matlab R2011b on a 2.5GHz Intel 2-Core processor (i.e., i5-3210m). |
| Experiment Setup | Yes | For our JS matching kernel k(M;h) JSM , we set the h as 10 and the greatest value of m as 40 (i.e., M = 40). For the WLSK kernel and the JTQK kernel, we set the highest dimension (i.e., the highest height of subtrees) of the Weisfeiler-Lehman isomorphism (for the WLSK kernel) and the tree-index method (for the JTQK kernel) as 10. For the DBMK kernel, we set the highest layer of the required DB representation as 10. |