Binary Matrix Factorisation via Column Generation
Authors: Reka A. Kovacs, Oktay Gunluk, Raphael A. Hauser3823-3831
AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on real world datasets demonstrate that our proposed method is effective at producing highly accurate factorisations and improves on the previously available best known results for 15 out of 24 problem instances. |
| Researcher Affiliation | Collaboration | R eka A. Kov acs,1 Oktay G unl uk, 2 Raphael A. Hauser 1 1 University of Oxford & The Alan Turing Institute 2 Cornell University |
| Pseudocode | No | The precise outline of the algorithm is given in Appendix 5.5. |
| Open Source Code | Yes | 1Appendix is available at arxiv.org/abs/2011.04457. |
| Open Datasets | Yes | In this section we present some experimental results with CG to demonstrate the practical applicability of our approach on eight real world categorical datasets that were downloaded from online repositories (Dua and Graff 2017), (Krebs 2008). |
| Dataset Splits | No | The paper does not provide specific details on training, validation, or test dataset splits (e.g., percentages, counts, or cross-validation setup). |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running the experiments. |
| Software Dependencies | Yes | Under exact pricing, the best heuristic solution is used as a warm start and PP is solved to optimality at each iteration using (CPLEX Optimization). ... CPLEX Optimization. 2018. Using the CPLEX Callable Library, V.12.8. |
| Experiment Setup | Yes | In Table 2 we present computational results comparing the optimality gap... under objective (6) using a 20 mins time budget. ... The exact details and parameters used in the computations can be found in Appendix 5.10.1. |