Safe Sample Screening for Robust Support Vector Machine
Authors: Zhou Zhai, Bin Gu, Xiang Li, Heng Huang6981-6988
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on a variety of benchmark datasets verify that our safe sample screening rules can significantly reduce the computational time. |
| Researcher Affiliation | Collaboration | Zhou Zhai,1 Bin Gu,1,2* Xiang Li,3 Heng Huang4 1School of Computer & Software, Nanjing University of Information Science & Technology, P.R.China 2JD Finance America Corporation 3Computer Science Department, University of Western Ontario, Canada 4Computer Engineering, University of Pittsburgh, USA |
| Pseudocode | Yes | Algorithm 1 Safe sample screening for single CIL problem; Algorithm 2 Safe sample screening for successive CIL problem |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described. It mentions LIBSVM as a source for datasets and for comparison, which is a third-party tool. |
| Open Datasets | Yes | Table 2: The benchmark datasets used in the experiments. Dateset Dimensionality Samples Source Cod RNA 8 59535 LIBSVM a9a 123 32561 LIBSVM letter 16 20000 LIBSVM ijcnn1 22 49990 LIBSVM; 1https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/. |
| Dataset Splits | No | The paper mentions selecting 20000 samples for training, but does not specify any training/validation/test dataset splits or percentages. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, processor types, memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper states: "We implement our algorithm in MATLAB." However, it does not provide specific version numbers for MATLAB or any other software libraries or dependencies. |
| Experiment Setup | Yes | The parameter C is selected from the set {0.1, 1, 10, 100}. The Gaussian kernel parameter κ is selected from the set {0.05, 0.5, 5}. The ramp loss function parameter s is fixed at 0. The optimization precision ϵ is set to be 10 8. We typically start sample screening after 50 iterations and screen the samples every 10 iterations. |