Proximity Operator of the Matrix Perspective Function and its Applications
Authors: Joong-Ho (Johann) Won
NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments verify that the evaluation of the proximity operator requires at most 8 Newton steps, taking less than 5s for 2000 by 2000 matrices on a standard laptop. Using this routine as a building block, we demonstrate the usefulness of the studied proximity operator in constrained maximum likelihood estimation of Gaussian mean and covariance, peudolikelihoodbased graphical model selection, and a matrix variant of the scaled lasso problem. |
| Researcher Affiliation | Academia | Joong-Ho Won Department of Statistics Seoul National University wonj@stats.snu.ac.kr |
| Pseudocode | Yes | Algorithm 1 Guarded Newton |
| Open Source Code | No | The paper mentions implementation in Julia but does not provide any concrete access information (specific repository link, explicit code release statement, or code in supplementary materials) for the source code of the described methodology. |
| Open Datasets | No | The paper describes generating synthetic data for its experiments (e.g., 'An N p data matrix X was sampled from independent Gaussian') rather than using publicly available datasets with concrete access information. |
| Dataset Splits | No | The paper describes generating synthetic data for its experiments, such as 'An N p data matrix X was sampled from independent Gaussian,' but does not specify training, validation, or testing dataset splits, exact percentages, or sample counts for reproduction. |
| Hardware Specification | Yes | The algorithm was implemented in the Julia programming language on a standard laptop (Macbook Pro 2019, i5@2.4GHz, 16GB RAM) |
| Software Dependencies | Yes | The algorithm was implemented in the Julia programming language on a standard laptop... and MOSEK was invoked via its Julia interface Convex.jl [33]. |
| Experiment Setup | No | The paper states, 'Detailed derivation of the PDHG iteration, setup, and convergence criteria for each problem appear in the Supplement,' indicating that specific experimental setup details, including hyperparameters, are not fully provided in the main text. |