Can Hybrid Geometric Scattering Networks Help Solve the Maximum Clique Problem?
Authors: Yimeng Min, Frederik Wenkel, Michael Perlmutter, Guy Wolf
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our empirical results demonstrate that our method outperforms representative GNN baselines in terms of solution accuracy and inference speed as well as conventional solvers like Gurobi with limited time budgets. |
| Researcher Affiliation | Academia | Yimeng Min Department of Computer Science Cornell University Ithaca, NY, USA min@cs.cornell.edu Frederik Wenkel Department of Mathematics and Statistics Université de Montréal Mila Quebec AI Institute Montreal, QC, Canada frederik.wenkel@umontreal.ca Michael A. Perlmutter Department of Mathematics University of California Los Angeles, CA, USA perlmutter@ucla.edu Guy Wolf Department of Mathematics and Statistics Université de Montréal Mila Quebec AI Institute Montreal, QC, Canada guy.wolf@umontreal.ca |
| Pseudocode | Yes | Algorithm 1 Maximum Clique Decoder |
| Open Source Code | No | The paper does not include an unambiguous statement or a direct link to a source-code repository for the methodology described. |
| Open Datasets | Yes | We evaluate our model on three popular real-world graph learning datasets, namely IMDB, COLLAB [Yanardag and Vishwanathan, 2015] and TWITTER [Yan et al., 2008]. |
| Dataset Splits | No | We note that setting the hyperparameters of the decoder requires some prior knowledge of the largest sizes of the maximum cliques present in the data, which we tune using the validation set. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | Yes | We further report results for the integerprogramming method Gurobi 9.0 [Gurobi Optimization, LLC, 2022]. |
| Experiment Setup | Yes | In the decoder (Section 3.5), we set κ equal to 1, 1 and 10 for the IMDB, COLLAB and TWITTER dataset, respectively. |