{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/safe-ml-surrogate-assisted-feature-extraction","title":"SAFE ML: Surrogate Assisted Feature Extraction for Model Learning","arxiv_id":"1902.11035","date":"2019-02-28","proceeding":null,"authors":["Alicja Gosiewska","Aleksandra Gacek","Piotr Lubon","Przemyslaw Biecek"],"abstract":"Complex black-box predictive models may have high accuracy, but opacity\ncauses problems like lack of trust, lack of stability, sensitivity to concept\ndrift. On the other hand, interpretable models require more work related to\nfeature engineering, which is very time consuming. Can we train interpretable\nand accurate models, without timeless feature engineering? In this article, we\nshow a method that uses elastic black-boxes as surrogate models to create a\nsimpler, less opaque, yet still accurate and interpretable glass-box models.\nNew models are created on newly engineered features extracted/learned with the\nhelp of a surrogate model. We show applications of this method for model level\nexplanations and possible extensions for instance level explanations. We also\npresent an example implementation in Python and benchmark this method on a\nnumber of tabular data sets.","url_abs":"http://arxiv.org/abs/1902.11035v1","url_pdf":"http://arxiv.org/pdf/1902.11035v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"safe-ml-surrogate-assisted-feature-extraction","repo_url":"https://github.com/olagacek/SAFE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"safe-ml-surrogate-assisted-feature-extraction","repo_url":"https://github.com/ModelOriented/SAFE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"safe-ml-surrogate-assisted-feature-extraction","repo_url":"https://github.com/ModelOriented/rSAFE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"safe-ml-surrogate-assisted-feature-extraction","repo_url":"https://github.com/jim-schwoebel/allie","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"automl","task_name":"AutoML"},{"task_slug":"feature-engineering","task_name":"Feature Engineering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}