Blended Matching Pursuit
Authors: Cyrille Combettes, Sebastian Pokutta
NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We implement BMP in Python 3 along with GMP and OMP [Locatello et al., 2017], the Accelerated Matching Pursuit algorithm (acc MP) [Locatello et al., 2018], and the Blended Conditional Gradients (BCG) [Braun et al., 2019] and Conditional Gradient with Enhancement and Truncation (Co GEn T) [Rao et al., 2015] algorithms for completeness. All algorithms share the same code framework to ensure fair comparison. No enhancement beyond basic coding was performed. We ran the experiments on a laptop under Linux Ubuntu 18.04 with Intel Core i7 3.5GHz CPU and 8GB RAM. |
| Researcher Affiliation | Academia | Cyrille W. Combettes Georgia Institute of Technology Atlanta, GA, USA cyrille@gatech.edu Sebastian Pokutta Zuse Institute Berlin and TU Berlin Berlin, Germany pokutta@zib.de |
| Pseudocode | Yes | Algorithm 1 Generalized/Orthogonal Matching Pursuit (GMP/OMP) |
| Open Source Code | No | The paper mentions implementing algorithms and a shared code framework ('All algorithms share the same code framework'), but it does not explicitly state that the source code for their proposed method (BMP) is publicly available or provide a link to a repository. |
| Open Datasets | No | The paper describes generating random data from Gaussian distributions and setting parameters (e.g., 'The random data are drawn from Gaussian distributions. For GMP, OMP, BCG, and Co GEn T, we represented the dual gaps by minv D f(xt), v , yielding a zig-zag plot dissimilar to the stair-like plot of the dual gap estimates |φt| of BMP. The Appendix contains additional experiments.'), but it does not provide access information (link, citation) to a publicly available dataset for training. |
| Dataset Splits | No | The paper states, 'We used early stopping on a validation set and present the test error...', indicating the use of a validation set. However, it does not provide specific details on the split percentages, sample counts, or the methodology for creating this split that would allow for reproduction. |
| Hardware Specification | Yes | We ran the experiments on a laptop under Linux Ubuntu 18.04 with Intel Core i7 3.5GHz CPU and 8GB RAM. |
| Software Dependencies | No | The paper mentions 'We implemented BMP in Python 3', but it does not specify version numbers for any libraries, frameworks, or other ancillary software components used, which are necessary for reproducible dependency information. |
| Experiment Setup | Yes | In BMP, we set κ = τ = 2 and we chose η = 5; see Appendix A.1 for an in-depth sensitivity analysis of BMP with respect to η. We did not perform any additional correction of the active sets (Line 22). |