Measures of Model Interpretability for Model Selection

Abstract : The literature lacks definitions for quantitative measures of model interpretability for automatic model selection to achieve high accuracy and interpretability, hence we define inherent model interpretability. We extend the work of Lipton et al. and Liu et al. from qualitative and subjective concepts of model interpretability to objective criteria and quantitative measures. We also develop another new measure called simplicity of sensitivity and illustrate prior, initial and posterior measurement. Measures are tested and validated with some measures recommended for use. It is demonstrated that high accuracy and high interpretability are jointly achievable with little to no sacrifice in either.
Complete list of metadatas

Cited literature [57 references]  Display  Hide  Download

https://hal.inria.fr/hal-02060060
Contributor : Hal Ifip <>
Submitted on : Thursday, March 7, 2019 - 10:37:41 AM
Last modification on : Friday, March 8, 2019 - 1:23:49 AM
Long-term archiving on: Monday, June 10, 2019 - 3:25:48 PM

File

 Restricted access
To satisfy the distribution rights of the publisher, the document is embargoed until : 2021-01-01

Please log in to resquest access to the document

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

André Carrington, Paul Fieguth, Helen Chen. Measures of Model Interpretability for Model Selection. 2nd International Cross-Domain Conference for Machine Learning and Knowledge Extraction (CD-MAKE), Aug 2018, Hamburg, Germany. pp.329-349, ⟨10.1007/978-3-319-99740-7_24⟩. ⟨hal-02060060⟩

Share

Metrics

Record views

52