Skip to Main content Skip to Navigation
Conference papers

Rule Extraction with Guaranteed Fidelity

Abstract : This paper extends the conformal prediction framework to rule extraction, making it possible to extract interpretable models from opaque models in a setting where either the infidelity or the error rate is bounded by a predefined significance level. Experimental results on 27 publicly available data sets show that all three setups evaluated produced valid and rather efficient conformal predictors. The implication is that augmenting rule extraction with conformal prediction allows extraction of models where test set errors or test sets infidelities are guaranteed to be lower than a chosen acceptable level. Clearly this is beneficial for both typical rule extraction scenarios, i.e., either when the purpose is to explain an existing opaque model, or when it is to build a predictive model that must be interpretable.
Document type :
Conference papers
Complete list of metadata

Cited literature [13 references]  Display  Hide  Download

https://hal.inria.fr/hal-01391055
Contributor : Hal Ifip <>
Submitted on : Wednesday, November 2, 2016 - 5:18:43 PM
Last modification on : Thursday, March 5, 2020 - 5:41:12 PM
Long-term archiving on: : Friday, February 3, 2017 - 2:42:03 PM

File

978-3-662-44722-2_30_Chapter.p...
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Ulf Johansson, Rikard König, Henrik Linusson, Tuve Löfström, Henrik Boström. Rule Extraction with Guaranteed Fidelity. 10th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Sep 2014, Rhodes, Greece. pp.281-290, ⟨10.1007/978-3-662-44722-2_30⟩. ⟨hal-01391055⟩

Share

Metrics

Record views

177

Files downloads

276