We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: From unbiased MDI Feature Importance to Explainable AI for Trees

Abstract: We attempt to give a unifying view of the various recent attempts to (i) improve the interpretability of tree-based models and (ii) debias the the default variable-importance measure in random Forests, Gini importance. In particular, we demonstrate a common thread among the out-of-bag based bias correction methods and their connection to local explanation for trees. In addition, we point out a bias caused by the inclusion of inbag data in the newly developed explainable AI for trees algorithms.
Comments: arXiv admin note: text overlap with arXiv:2003.02106
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Computation (stat.CO)
Cite as: arXiv:2003.12043 [stat.ML]
  (or arXiv:2003.12043v4 [stat.ML] for this version)

Submission history

From: Markus Loecher [view email]
[v1] Thu, 26 Mar 2020 17:16:58 GMT (114kb,D)
[v2] Wed, 15 Apr 2020 17:45:50 GMT (122kb,D)
[v3] Sun, 24 May 2020 15:44:32 GMT (127kb,D)
[v4] Thu, 30 Sep 2021 14:35:53 GMT (126kb,D)

Link back to: arXiv, form interface, contact.