{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/consistent-feature-attribution-for-tree","title":"Consistent feature attribution for tree ensembles","arxiv_id":"1706.06060","date":"2017-06-19","proceeding":null,"authors":["Scott M. Lundberg","Su-In Lee"],"abstract":"Note that a newer expanded version of this paper is now available at:\narXiv:1802.03888\n  It is critical in many applications to understand what features are important\nfor a model, and why individual predictions were made. For tree ensemble\nmethods these questions are usually answered by attributing importance values\nto input features, either globally or for a single prediction. Here we show\nthat current feature attribution methods are inconsistent, which means changing\nthe model to rely more on a given feature can actually decrease the importance\nassigned to that feature. To address this problem we develop fast exact\nsolutions for SHAP (SHapley Additive exPlanation) values, which were recently\nshown to be the unique additive feature attribution method based on conditional\nexpectations that is both consistent and locally accurate. We integrate these\nimprovements into the latest version of XGBoost, demonstrate the\ninconsistencies of current methods, and show how using SHAP values results in\nsignificantly improved supervised clustering performance. Feature importance\nvalues are a key part of understanding widely used models such as gradient\nboosting trees and random forests, so improvements to them have broad practical\nimplications.","url_abs":"http://arxiv.org/abs/1706.06060v6","url_pdf":"http://arxiv.org/pdf/1706.06060v6.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":"consistent-feature-attribution-for-tree","repo_url":"https://github.com/bgreenwell/fastshap","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"feature-importance","task_name":"Feature Importance"}],"methods":[{"method_slug":"shap","method_name":"SHAP"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1706.06060","atlas_url":"https://app.syntology.ai/?focus=1706.06060","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"1706.06060"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"1706.06060\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"1706.06060"},"call":"get_concepts_for_paper(arxiv_id=\"1706.06060\")"}],"by_repo":[{"repo":"bgreenwell/fastshap","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"ea8415501874fbc06fbffaffc2dfc59a49d25427aaff8ff2986881b12ff37cac","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}