Sparse Subspace Clustering with Entropy-Norm
Authors: Liang Bai, Jiye Liang
ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally, we provide the experimental analysis to compare the efficiency and effectiveness of sparse subspace clustering and spectral clustering on ten benchmark data sets. The theoretical and experimental analysis can well help users for the selection of highdimensional data clustering algorithms. |
| Researcher Affiliation | Academia | Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education, School of Computer and Information Technology, Shanxi University, Taiyuan, Shanxi Province, China. Correspondence to: Jiye Liang <ljy@sxu.edu.cn>. |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The codes of these algorithms have been publicly shared by their authors. (This refers to SSC, CAN, and SSC+E as existing algorithms, not new code released by the authors of *this* paper for their specific methodology if it were novel.) |
| Open Datasets | Yes | We carry out these algorithms on 10 benchmark data sets (Bache & Lichman; Cai) which are described in Table 1. |
| Dataset Splits | No | The paper does not explicitly mention separate training, validation, and test splits with specific details. It mentions testing algorithms on benchmark datasets and setting the number of clusters to the true number of classes, but no detailed split information including validation. |
| Hardware Specification | Yes | The experiments are conducted on an Intel i9-7940X CPU@3.10HZ and 128G RAM. |
| Software Dependencies | No | The paper does not provide specific version numbers for software dependencies or libraries used. |
| Experiment Setup | Yes | Before the comparisons, we need to set some parameters for these algorithms as follows. We set the number of clusters k is equal to its true number of classes on each of the given data sets. For the parameter λ, we test each algorithm with different λ values which are selected in the set {λ1 = δ 50, λ2 = δ 40, λ3 = δ 30, λ4 = δ 20, λ5 = δ 10, λ6 = δ}, where δ is the covariance of a data set. Besides, the SSC and CAN algorithms need to set the number of the nearest neighbors K. We set K to 10 in our experiments. |