Accelerated Projected Gradient Algorithms for Sparsity Constrained Optimization Problems
Authors: Jan Harold Alcantara, Ching-pei Lee
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments demonstrate that the proposed accelerated algorithms are magnitudes faster than their non-accelerated counterparts as well as the state of the art. |
| Researcher Affiliation | Academia | Jan Harold Alcantara Academia Sinica Taipei, Taiwan jan.harold.alcantara@gmail.com Ching-pei Lee Academia Sinica Taipei, Taiwan leechingpei@gmail.com |
| Pseudocode | Yes | Algorithm 1: Accelerated projected gradient algorithm by extrapolation (APG) is presented on page 7. Algorithm 2: Accelerated projected gradient algorithm by subspace identification (PG+) is presented on page 8. |
| Open Source Code | No | The paper does not provide any explicit statement about releasing the source code for the described methodology or a link to a code repository. |
| Open Datasets | Yes | The algorithms are implemented in MATLAB and tested with public datasets in Tables 2 and 3 in Appendix B. |
| Dataset Splits | No | The paper mentions using "test data" for evaluation but does not provide specific details on how the datasets were split into training, validation, or test sets, such as percentages, sample counts, or a specific splitting methodology. |
| Hardware Specification | No | The paper does not specify any particular hardware components such as GPU models, CPU types, or cloud computing instance details used for running the experiments. |
| Software Dependencies | No | The paper mentions that "The algorithms are implemented in MATLAB" but does not specify a version number or any other software dependencies with their versions. |
| Experiment Setup | Yes | All algorithms compared start from w0 = 0 and terminate when the first-order optimality condition Residual(w) := w PAs (w λ f (w)) /(1 + w + λ f (w) ) < ˆϵ (24) is met for some given ˆϵ > 0. More setting and parameter details of our experiments are in Appendix B. |