OptimalFeatureSelection Documentation

OptimalFeatureSelection contains learners for conducting optimal feature selection for classification and regression problems. This documentation includes:

• quick start guides for each of the problem types that contain a demo of OptimalFeatureSelection in action:
  • classification
  • regression
• an introduction to the optimal feature selection problem and the learners used to tackle this problem, including descriptions of the available parameters
• a guide to the available options for visualizing optimal feature selection
• an example for handling missing data in regression
• a guide to coordinated-sparsity fitting
• the OptimalFeatureSelection API reference

Citing OptimalFeatureSelection

If you use Optimal Feature Selection in your work, we kindly ask that you cite the Interpretable AI software modules. We also ask that you reference the original work that first introduced the relevant algorithm:

• Optimal Feature Selection for Classification:

  @article{bertsimas2017sparse,
    title={Sparse classification and phase transitions: A discrete optimization perspective},
    author={Bertsimas, Dimitris and Pauphilet, Jean and Van Parys, Bart},
    journal={arXiv preprint arXiv:1710.01352},
    year={2017}
  }

• Optimal Feature Selection for Regression (relaxation):

  @article{bertsimas2019sparse,
    title={Sparse Regression: Scalable algorithms and empirical performance},
    author={Bertsimas, Dimitris and Pauphilet, Jean and Van Parys, Bart},
    journal={arXiv preprint arXiv:1902.06547},
    year={2019}
  }

• Optimal Feature Selection for Regression (exact):

  @article{bertsimas2017sparse,
    title={Sparse high-dimensional regression: Exact scalable algorithms and phase transitions},
    author={Bertsimas, Dimitris and Van Parys, Bart},
    journal={arXiv preprint arXiv:1709.10029},
    year={2017}
  }

• Coordinated-Sparsity Fitting:

  @article{bertsimas2021sparse,
    title={Sparse regression over clusters: SparClur},
    author={Bertsimas, Dimitris and Dunn, Jack and Kapelevich, Lea and Zhang, Rebecca},
    journal={Optimization Letters},
    pages={1--16},
    year={2021},
    publisher={Springer}
  }