{"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/model-agnostic-supervised-local-explanations","title":"Model Agnostic Supervised Local Explanations","arxiv_id":"1807.02910","date":"2018-07-09","proceeding":"NeurIPS 2018 12","authors":["Gregory Plumb","Denali Molitor","Ameet Talwalkar"],"abstract":"Model interpretability is an increasingly important component of practical\nmachine learning. Some of the most common forms of interpretability systems are\nexample-based, local, and global explanations. One of the main challenges in\ninterpretability is designing explanation systems that can capture aspects of\neach of these explanation types, in order to develop a more thorough\nunderstanding of the model. We address this challenge in a novel model called\nMAPLE that uses local linear modeling techniques along with a dual\ninterpretation of random forests (both as a supervised neighborhood approach\nand as a feature selection method). MAPLE has two fundamental advantages over\nexisting interpretability systems. First, while it is effective as a black-box\nexplanation system, MAPLE itself is a highly accurate predictive model that\nprovides faithful self explanations, and thus sidesteps the typical\naccuracy-interpretability trade-off. Specifically, we demonstrate, on several\nUCI datasets, that MAPLE is at least as accurate as random forests and that it\nproduces more faithful local explanations than LIME, a popular interpretability\nsystem. Second, MAPLE provides both example-based and local explanations and\ncan detect global patterns, which allows it to diagnose limitations in its\nlocal explanations.","url_abs":"http://arxiv.org/abs/1807.02910v3","url_pdf":"http://arxiv.org/pdf/1807.02910v3.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":"model-agnostic-supervised-local-explanations","repo_url":"https://github.com/GDPlumb/MAPLE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"model-agnostic-supervised-local-explanations","repo_url":"https://github.com/IBCNServices/skMAPLE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"feature-selection","task_name":"feature selection"},{"task_slug":"model","task_name":"model"}],"methods":[{"method_slug":"interpretability","method_name":"Interpretability"},{"method_slug":"lime","method_name":"LIME"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.02910","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}