Conic Optimization for Quadratic Regression Under Sparse Noise
Authors: Igor Molybog, Ramtin Madani, Javad Lavaei
JMLR 2020 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The efficacy of the developed methods is demonstrated in different case studies, including data analytics for a European power grid. |
| Researcher Affiliation | Academia | Igor Molybog EMAIL Department of Industrial Engineering and Operations Research University of California Berkeley, CA 94720, USA Ramtin Madani EMAIL Department of Electrical Engineering University of Texas Arlington, TA 76102, USA Javad Lavaei EMAIL Department of Industrial Engineering and Operations Research University of California Berkeley, CA 94720, USA |
| Pseudocode | Yes | Algorithm 1 Conic Hard Thresholding |
| Open Source Code | No | The paper does not provide explicit statements about releasing source code or a link to a code repository. It mentions using 'MOSEK v7. SOCP-solving procedure' but this is a third-party tool. |
| Open Datasets | Yes | The experiment is run on the PEGASE 1354-bus European system borrowed from the MATPOWER package (Fliscounakis et al., 2013; Josz et al., 2016). We use the IEEE 300-bus benchmark system from the MATPOWER package |
| Dataset Splits | No | The paper mentions how corrupted measurements are chosen or the percentage of corrupted measurements, but does not provide specific training/test/validation dataset splits (e.g., percentages, sample counts, or predefined splits) for reproducibility. |
| Hardware Specification | Yes | The results of this part are produced using the standard MOSEK v7. SOCP-solving procedure, run in MATLAB on a 12-core 2.2GHz machine with 256GB RAM. |
| Software Dependencies | Yes | The results of this part are produced using the standard MOSEK v7. SOCP-solving procedure, run in MATLAB |
| Experiment Setup | Yes | The parameter µ is chosen as 10 2. Regarding Algorithm 1, the parameter k is selected as the true number of corrupted measurements, the tolerance ε is set to 10 3, and the algorithm is terminated early if the number of conic iterations exceeds 50. |