GridEx: An Algorithm for Knowledge Extraction from Black-Box Regressors


Federico Sabbatini, Giovanni Ciatto, Andrea Omicini

Explainable and Transparent AI and Multi-Agent Systems. Third International Workshop, EXTRAAMAS 2021, Virtual Event, May 3–7, 2021, Revised Selected Papers, pages 18-38
Lecture Notes in Computer Science (Lecture Notes in Artificial Intelligence) 12688, July 2021
Springer, Cham, Switzewland
Davide Calvaresi, Amro Najjar, Michael Winikoff, Kary Främling (eds.)

Knowledge-extraction methods are applied to ML-based predictors to attain explainable representations of their operation when the lack of interpretable results constitutes a problem. Several algorithms have been proposed for knowledge extraction, mostly focusing on the extraction of either lists or trees of rules. Yet, most of them only support supervised learning – and, in particular, classification – tasks. Iter is among the few rule-extraction methods capable of extracting symbolic rules out of sub-symbolic regressors. However, its performance – here intended as the interpretability of the rules it extracts – easily degrades as the complexity of the regression task at hand increases.

In this paper we propose GridEx, an extension of the Iter algorithm, aimed at extracting symbolic knowledge – in the form of lists of if-then-else rules – from any sort of sub-symbolic regressor—there including neural networks of arbitrary depth. With respect to Iter, GridEx produces shorter rule lists retaining higher fidelity w.r.t. the original regressor. We report several experiments assessing GridEx performance against Iter and Cart (i.e., decision-tree regressors) used as benchmarks.

(keywords) Explainable AI; Knowledge extraction; Interpretable prediction; Regression; Iter; GridEx 
 @incollection{gridex-extraamas2021,
address = {Cham, Switzerland},
author = {Sabbatini, Federico and Ciatto, Giovanni and Omicini, Andrea},
booktitle = {Explainable and Transparent AI and Multi-Agent Systems. Third International Workshop, EXTRAAMAS 2021, Virtual Event, May 3--7, 2021, Revised Selected Papers},
dblp = {conf/atal/SabbatiniCO21},
doi = {10.1007/978-3-030-82017-6_2},
editor = {Calvaresi, Davide and Najjar, Amro and Winikoff, Michael and Fr{\"a}mling, Kary},
iris = {11585/834616},
isbn = {978-3-030-82016-9},
isbn-online = {978-3-030-82017-6},
issn = {0302-9743},
issn-online = {1611-3349},
pages = {18--38},
publisher = {Springer Nature},
scopus = {2-s2.0-85113335454},
series = {Lecture Notes in Computer Science},
subseries = {Lecture Notes in Artificial Intelligence},
title = {{GridEx}: An Algorithm for Knowledge Extraction from Black-Box Regressors},
url = {http://link.springer.com/10.1007/978-3-030-82017-6_2},
volume = 12688,
year = 2021

Talks

Journals & Series

Events

  • EXplainable and TRAnsparent AI and Multi-Agent Systems: Third International Workshop (EXTRAAMAS 2021) — 03/05/2021–04/05/2021

Publication

— authors

Federico Sabbatini, Giovanni Ciatto, Andrea Omicini

— editors

Davide Calvaresi, Amro Najjar, Michael Winikoff, Kary Främling

— status

published

— sort

paper in proceedings

Venue

— volume

Explainable and Transparent AI and Multi-Agent Systems. Third International Workshop, EXTRAAMAS 2021, Virtual Event, May 3–7, 2021, Revised Selected Papers

— series

Lecture Notes in Computer Science / Lecture Notes in Artificial Intelligence

— volume

12688

— pages

18-38

— publication date

July 2021

URLs

original page  |  original PDF  |  open access PDF

Identifiers

— DOI

10.1007/978-3-030-82017-6_2

— DBLP

conf/atal/SabbatiniCO21

— IRIS

11585/834616

— Scholar

855045469053426346

— Scopus

2-s2.0-85113335454

— WoS / ISI

000691781800002

— print ISSN

0302-9743

— online ISSN

1611-3349

— print ISBN

978-3-030-82016-9

— online ISBN

978-3-030-82017-6

BibTeX

— BibTeX ID
gridex-extraamas2021
— BibTeX category
incollection

Files

Open Access PDF

View this PDF full screen

You do not have the plugin required to display this PDF file. You can still download it: extraamas-2021-iter.pdf

Partita IVA: 01131710376 - Copyright © 2008-2022 APICe@DISI Research Group - PRIVACY