Simultaneous Graph Signal Clustering and Graph Learning
Authors: Abdullah Karaaslanli, Selin Aviyente
ICML 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | results on simulated and real data indicate the effectiveness of the proposed method. |
| Researcher Affiliation | Academia | Abdullah Karaaslanli 1 Selin Aviyente 1 1Department of Electrical and Computer Engineering, Michigan State University, East Lansing, MI, US. |
| Pseudocode | Yes | Algorithm 1 GS Clustering with Simultaneous GL ... Algorithm 2 Initialization Procedure |
| Open Source Code | Yes | 1Codes are available at the following github repository: https://github.com/SPLab-aviyente/GRASCale |
| Open Datasets | Yes | In this section, the proposed method is applied to a real world data clustering problem, where the aim is to cluster the digits of MNIST dataset... We selected 400 images corresponding to digits 0, 1, 2 and 3. |
| Dataset Splits | No | The paper mentions running algorithms multiple times and averaging performance, but it does not specify explicit training, validation, or test dataset splits (e.g., percentages or sample counts) for the datasets used. |
| Hardware Specification | No | The paper does not explicitly describe the specific hardware (e.g., GPU models, CPU types) used to run the experiments. |
| Software Dependencies | No | The paper mentions the use of certain algorithms and models (e.g., SC, GLMM, K-Graphs) but does not provide specific version numbers for any software libraries or dependencies used in the implementation. |
| Experiment Setup | Yes | We provided the ground truth k as an input to all methods. ... For all methods, we set this hyperparameter to a value that results in graphs with sparsity levels between 0.1 and 0.152. ... For GRASCale, we set b = 9... Finally, SC is applied to a binary k-nearest neighbor graph with the number of neighbors set to 5. The same graph is used as Lc for the proposed method. ... We set the maximum number of iterations for each run to a small number, e.g., 100. ... we set α1 = 10 in all of our data analysis without any fine-tuning. |