Subset Selection by Pareto Optimization with Recombination

Authors: Chao Qian, Chao Bian, Chao Feng2408-2415

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on unsupervised feature selection and sparse regression show the superiority of PORSS over POSS.
Researcher Affiliation Academia 1National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210023, China 2School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China qianc@lamda.nju.edu.cn, chaobian12@gmail.com, chaofeng@mail.ustc.edu.cn
Pseudocode Yes Algorithm 1 POSS Algorithm; Algorithm 2 PORSS Algorithm
Open Source Code No The paper provides links to datasets used in the experiments but does not provide specific links or statements regarding the open-sourcing of its own methodology's code.
Open Datasets Yes https://archive.ics.uci.edu/ml/datasets.html, https://www.csie. ntu.edu.cn/ cjlin/libsvmtools/datasets/ and http://www.cl.cam.ac. uk/research/dtg/attarchive/facedatabase.html. The paper lists datasets such as sonar, phishing, Hill-Valley, mediamill, musk, CT-slices, ISOLET, mnist, SVHN, ORL, svmguide3, triazines, clean1, usps, scene, protein, colon-cancer, cifar10, leukemia, small NORB.
Dataset Splits No The paper does not provide specific details on dataset splits (e.g., percentages, sample counts for training, validation, or testing sets), nor does it mention cross-validation. It only states that runs were repeated ten times.
Hardware Specification No The paper does not provide any specific hardware details such as GPU/CPU models, processor types, or memory amounts used for running the experiments.
Software Dependencies No The paper mentions algorithms and models like NSGA-II and Lasso but does not provide specific software dependencies (e.g., libraries, frameworks) with version numbers needed to replicate the experiments.
Experiment Setup Yes As suggested in (Qian, Yu, and Zhou 2015), the number T of iterations of POSS is set to 2ek2n. Note that POSS in Algorithm 1 requires one objective evaluation for the newly generated solution x in each iteration, whereas PORSS in Algorithm 2 needs to evaluate two new solutions x , y . For the fairness of comparison, the number T of iterations of PORSS is set to ek2n; thus, the same number of objective evaluations is used. The budget k is set to 8.