{"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/dalex-explainers-for-complex-predictive","title":"DALEX: explainers for complex predictive models","arxiv_id":"1806.08915","date":"2018-06-23","proceeding":null,"authors":["Przemyslaw Biecek"],"abstract":"Predictive modeling is invaded by elastic, yet complex methods such as neural\nnetworks or ensembles (model stacking, boosting or bagging). Such methods are\nusually described by a large number of parameters or hyper parameters - a price\nthat one needs to pay for elasticity. The very number of parameters makes\nmodels hard to understand. This paper describes a consistent collection of\nexplainers for predictive models, a.k.a. black boxes. Each explainer is a\ntechnique for exploration of a black box model. Presented approaches are\nmodel-agnostic, what means that they extract useful information from any\npredictive method despite its internal structure. Each explainer is linked with\na specific aspect of a model. Some are useful in decomposing predictions, some\nserve better in understanding performance, while others are useful in\nunderstanding importance and conditional responses of a particular variable.\nEvery explainer presented in this paper works for a single model or for a\ncollection of models. In the latter case, models can be compared against each\nother. Such comparison helps to find strengths and weaknesses of different\napproaches and gives additional possibilities for model validation. Presented\nexplainers are implemented in the DALEX package for R. They are based on a\nuniform standardized grammar of model exploration which may be easily extended.\nThe current implementation supports the most popular frameworks for\nclassification and regression.","url_abs":"http://arxiv.org/abs/1806.08915v2","url_pdf":"http://arxiv.org/pdf/1806.08915v2.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":"dalex-explainers-for-complex-predictive","repo_url":"https://github.com/pbiecek/DALEX","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}