Tensorized Incomplete Multi-View Clustering with Intrinsic Graph Completion
Authors: Shuping Zhao, Jie Wen, Lunke Fei, Bob Zhang
AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on several realworld databases illustrates that the proposed method can outperform the other state-of-the-art related methods for incomplete multi-view clustering. |
| Researcher Affiliation | Academia | 1PAMI Research Group, Department of Computer and Information Science, University of Macau, Taipa, Macau 2Shenzhen Key Laboratory of Visual Object Detection and Recognition, Harbin Institute of Technology, Shenzhen, Shenzhen, 518055, China 3School of Computer Science, Guangdong University of Technology, Guangzhou, 510006, China. |
| Pseudocode | Yes | Algorithm 1: The proposed TIMVC IGC |
| Open Source Code | No | The paper mentions several datasets with URLs but does not provide a link or explicit statement about the availability of its own source code. |
| Open Datasets | Yes | Handwritten Multi-feature Dataset1 (Handwritten) includes 10 classes, i.e., digits 0-9 , where each class contains 200 handwritten samples. ... 1https://archive.ics.uci.edu/ml/datasets/Multiple Features. The Columbia Object Image Library2 (COIL-20) totally includes 1,440 images from 20 classes. ... 2http://www.cs.columbia.edu/CAVE/software/softlib/coil20.php Caltech101 database totally consists of 101 objects... Referring to (Zhao, Liu, and Fu 2016), a subset of the BUAA-visnir face database 3 (BUAA)... 3https://github.com/hdzhao/IMG/tree/master/data. |
| Dataset Splits | No | The paper describes how incomplete data was constructed (e.g., 'randomly removing p%...'), but does not specify distinct training, validation, or testing splits for model development or evaluation in the conventional sense. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used to run the experiments (e.g., CPU, GPU models, memory). |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers. It mentions the use of 'eigs in (Wright and Trefethen 2001)' for acceleration, but this is a function reference, not a full software stack specification with versions. |
| Experiment Setup | Yes | Three parameters, i.e., λ1, λ2, and λ3 need to be adjusted in Algorithm 1. ... It is obvious that our method is insensitive to λ1 in the range of [10 5, 102]. In addition, it can be seen that the highest clustering accuracy can be guaranteed for all databases with λ2 [102, 105] and λ3 = 103. |