Shapley Values Infidelity

Daniel M. Ph.D. Tom

No 2p5d3, OSF Preprints from Center for Open Science

Abstract: Shapley values attempt to explain ML models using flat additive factors disregarding any tree hierarchy, and fails to distinguish between two different trees. We have been using log odds for segmentation tree of logistic regression models. Log odds faithfully reflect tree hierarchy and therefore explain decision trees, forests, and GBM much better.

Date: 2023-06-06
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://osf.io/download/647f33e8ed6d8200cf7b357f/

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:osf:osfxxx:2p5d3

DOI: 10.31219/osf.io/2p5d3

Access Statistics for this paper

More papers in OSF Preprints from Center for Open Science
Bibliographic data for series maintained by OSF ().