Consistent and Specific Multi-View Subspace Clustering
Authors: Shirui Luo, Changqing Zhang, Wei Zhang, Xiaochun Cao
AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental evaluations on four benchmark datasets demonstrate that the proposed approach achieves better performance over several state-of-the-arts. [...] Extensive experiments on benchmark datasets demonstrate that our model outperforms several baseline methods and state-of-the-art methods. |
| Researcher Affiliation | Academia | Shirui Luo,1,2 Changqing Zhang,3 Wei Zhang,1 Xiaochun Cao1,2 1SKLOIS, Institute of Information Engineering, Chinese Academy of Sciences 2School of Cyber Security, University of Chinese Academy of Sciences 3School of Computer Science and Technology, Tianjin University, Tianjin, China |
| Pseudocode | Yes | Algorithm 1 CSMSC: Consistent and Specific Multi-view Subspace Clustering |
| Open Source Code | No | The paper does not provide any concrete access information (e.g., a link to a repository) or an explicit statement about releasing the source code. |
| Open Datasets | Yes | Four benchmark datasets are adopted in our evaluation. Yale is a widely used face dataset [...] Notting-Hill video face dataset [...] ORL face dataset [...] BBCSport (Xia et al. 2014) contains 544 documents from the BBC Sport website |
| Dataset Splits | No | The paper states 'we tune the parameter λC and λD in the range of (0,1] and report the best performing results' which implies some validation, but it does not specify the dataset splits (e.g., percentages, counts, or standard splits for these datasets) for training, validation, or testing. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., GPU/CPU models, memory) used to run the experiments. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers (e.g., Python 3.x, PyTorch 1.x, specific libraries or solvers) required to replicate the experiments. |
| Experiment Setup | Yes | In our experiments, we tune the parameter λC and λD in the range of (0,1] and report the best performing results. [...] For face datasets, we resize the images to 48 48 and extract three types of features: View1 intensity (4,096 dimensions), View2 LBP (Ojala, Pietikainen, and Maenpaa 2002) (3,304 dimensions) and View3 Gabor (Lades et al. 1993) (6,750 dimensions). [...] BBCSport dataset only has two views, View1: 3,183 dimensions and View2: 3,203 dimensions, respectively. |