Multiple Kernelk-Means Clustering with Matrix-Induced Regularization
Authors: Xinwang Liu, Yong Dou, Jianping Yin, Lei Wang, En Zhu
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | As experimentally demonstrated on five challenging MKL benchmark data sets, our algorithm significantly improves existing MKKM and consistently outperforms the state-of-the-art ones in the literature, verifying the effectiveness and advantages of incorporating the proposed matrix-induced regularization. |
| Researcher Affiliation | Academia | Xinwang Liu, Yong Dou, Jianping Yin School of Computer National University of Defense Technology Changsha, China, 410073 Lei Wang School of Computer Science and Software Engineering University of Wollongong NSW, Australia, 2522 En Zhu School of Computer National University of Defense Technology Changsha, China, 410073 |
| Pseudocode | Yes | Our algorithm for solving Eq.(8) is outlined in Algorithm 1 |
| Open Source Code | No | The Matlab codes of KKM, MKKM and LMKKM are publicly available from https://github.com/mehmetgonen/ lmkkmeans. For RMKKM, CRSC, RMSC and RCE, we download their matlab implementations from authors websites and use them for comparison in our experiments. |
| Open Datasets | Yes | We evaluate the clustering performance of our algorithm on five MKL benchmark data sets, including Oxford Flower172, Protein fold prediction3, UCI-Digital4, Oxford Flower1025 and Caltech1026. For the other data sets, all kernel matrices are pre-computed and can be publicly downloaded from the above websites. |
| Dataset Splits | No | The paper mentions evaluating clustering performance on benchmark datasets and selecting parameters by grid search, but it does not specify explicit training/validation/test dataset splits with percentages or sample counts. |
| Hardware Specification | No | The paper does not specify the hardware (e.g., GPU/CPU models, memory, or specific computing environment) used for running the experiments. |
| Software Dependencies | No | The paper mentions 'Matlab codes' for comparison algorithms but does not provide specific version numbers for Matlab or any other software dependencies used in the experiments. |
| Experiment Setup | Yes | In all experiments, all base kernels are first centered and then scaled so that for all i and p we have Kp(xi, xi) = 1. For all data sets, we assume that the true number of clusters is known and we set it to be the true number of classes. For our proposed algorithm, its regularization parameter is chosen from [2 15, 2 14, , 215] by grid search. For all algorithms, we repeat each experiment for 50 times with random initialization to reduce the affect of randomness caused by k-means, and report the best result. |