Consistency of Multiple Kernel Clustering
Authors: Weixuan Liang, Xinwang Liu, Yong Liu, Chuan Ma, Yunping Zhao, Zhe Liu, En Zhu
ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally, extensive experiments are conducted to verify the theoretical results and the effectiveness of the proposed large-scale strategy. To verify the proposed theoretical results, we conduct experiments on commonly used datasets. The numerical experiments substantiate the correctness of the derived bounds. Moreover, we perform our algorithm on large-scale datasets to verify its effectiveness and efficiency. |
| Researcher Affiliation | Academia | 1College of Computer, National University of Defense Technology, Changsha, China 2Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China 3Beijing Key Laboratory of Big Data Management and Analysis Methods, Beijing, China 4Zhejiang Laboratory, Hangzhou, China. |
| Pseudocode | No | The paper describes the algorithm steps in numbered text format in Section 6, but it does not provide structured pseudocode or a formally labeled algorithm block. |
| Open Source Code | No | The paper does not provide any explicit statements about making the source code available or include a link to a code repository. |
| Open Datasets | Yes | Table 1. Benchmark datasets Datasets Samples Kernels Clusters Flo17 1360 7 17 Flo102 8189 4 102 DIGIT 2000 3 10 PFold 694 12 27 CCV 6773 3 20 Reuters 18758 5 6 Table 2. Large-scale datasets used in the experiments Dataset Samples Views Clusters NUSWIDE 30000 5 31 Aw A 30475 6 50 CIFAR10 50000 3 10 Yt Video 101499 5 31 Winnipeg 325834 2 7 Covertype 581012 2 10 The URLs of the datasets in Table 1 are as follows: ... The large-scale datasets in Table 2 can be downloaded from |
| Dataset Splits | No | The paper does not explicitly provide training/validation/test dataset splits, percentages, or cross-validation setup. |
| Hardware Specification | Yes | All experiments are conducted on a laptop with Intel(R) Core(TM)-i7-10870H CPU. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers. |
| Experiment Setup | Yes | For each view, we use the Gaussian RBF kernel to construct the kernel similarity matrix between the whole training set Sn and the selected subset Sr, i.e., K(x, z) = exp x z 2 / (2σ2), where x Sn, z Sr, and σ is the square root of the average interpoint distance between Sn and Sr, i.e., σ2 = 1/(nr) P n i=1 P r t=1 x t 2. As proven in Section 6, by setting r = Θ( n), the proposed algorithm can have a favorable statistical and computational trade-off. For a sufficient r, we let r = 3 n . |