Linearity-Aware Subspace Clustering

Authors: Yesong Xu, Shuo Chen, Jun Li, Jianjun Qian8770-8778

AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental Results" section, as well as discussing "Clustering Performance and Analysis" and presenting data in "Table 2: Performance comparison of all compared methods on the five benchmark datasets."
Researcher Affiliation Academia 1PCA Lab , Nanjing University of Science and Technology yesong xu@163.com, {shuochen, junli, csjqian}@njust.edu.cn
Pseudocode Yes Algorithm 1: Subspace Segmentation via LASC
Open Source Code No The paper does not provide an explicit statement or a link to open-source code for the described methodology.
Open Datasets Yes Datasets: In our experiments, five benchmark datasets are selected, including two handwritten digits datasets MNIST1 and USPS2, two object recognition datasets COIL1003 and CIFAR104, and an face image dataset Extended Yale B(Ex Yale B)5. (with URLs provided in footnotes)
Dataset Splits Yes For all above mentioned algorithms, the parameters are tuned by the cross validation technique to guarantee their possibly optimal performance.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments, such as CPU or GPU models, or memory specifications.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., Python version, specific libraries like PyTorch or TensorFlow with their versions) that would be needed to replicate the experiment.
Experiment Setup Yes Table 2 presents the clustering results and the tuned parameters of all tested approaches. For example, for SSC on MNIST with 6 classes, the parameters are '0.001, 0.1'. The paper also states 'Initialize: C0 = S0 = 0 and ε1 = 10 4, ε2 = 10 5' and mentions a 'maximal iteration number T'.