Projected Tensor Power Method for Hypergraph Community Recovery
Authors: Jinxin Wang, Yuen-Man Pun, Xiaolu Wang, Peng Wang, Anthony Man-Cho So
ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We also conduct numerical experiments to validate our theoretical findings. ... In this section, we report the recovery performance and numerical efficiency of our proposed PTPM for recovering communities on synthetic/real data. |
| Researcher Affiliation | Academia | 1Department of Systems Engineering and Engineering Management, The Chinese University of Hong Kong, Hong Kong 2CIICADA Lab, School of Engineering, Australian National University, Canberra 3Department of Electrical Engineering and Computer Science, University of Michigan, Ann Arbor. |
| Pseudocode | Yes | Algorithm 1 Projected Tensor Power Method for Solving Problem (MLE) |
| Open Source Code | No | The paper does not provide an explicit statement or link for open-source code for the methodology described. |
| Open Datasets | Yes | Based on the 1984 US Congressional voting records available at the UCI repository |
| Dataset Splits | No | The paper mentions generating synthetic hypergraphs and using the UCI dataset, but it does not specify explicit training, validation, or test dataset splits (e.g., percentages or sample counts). It describes experiment runs like 'run PTPM 8 times' or 'run each algorithm 10 times' but not data partitioning. |
| Hardware Specification | No | The paper mentions the use of MATLAB functions for computations but does not specify any hardware details like GPU models, CPU types, or memory used for the experiments. |
| Software Dependencies | Yes | We use the Tensor Toolbox (Kolda et al., 2017) to perform tensor operations and compute A h (Ht) (d 1)i based on (7). (Kolda, T. G., Bader, B. W., Acar Ataman, E. N., Dunlavy, D., Bassett, R., Battaglino, C. J., Plantenga, T., Chi, E., Hansen, S., and USDOE. Tensor toolbox for matlab v. 3.0, 3 2017. URL https://www.osti.gov/ /servlets/purl/1349514.) |
| Experiment Setup | Yes | We set n = 210, K = 3 in the experiments and let the parameter α vary from 0 to 120 with increments of 3 and the parameter β vary from 0 to 40 with increments of 1. For each pair of α and β, we generate 5 instances and calculate the success ratio of exact recovery for all the tested methods. ... In each hypergraph realization, we run PTPM 8 times with different random initial points and then plot the distances of the iterative points to the ground truth ... terminate PTPM when the iteration number reaches 20. |