Comparing Feature Importance and Rule Extraction for Interpretability on Text Data

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lopardo, Gianluigi, Garreau, Damien
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911222099607552
author Lopardo, Gianluigi
Garreau, Damien
author_facet Lopardo, Gianluigi
Garreau, Damien
contents Complex machine learning algorithms are used more and more often in critical tasks involving text data, leading to the development of interpretability methods. Among local methods, two families have emerged: those computing importance scores for each feature and those extracting simple logical rules. In this paper we show that using different methods can lead to unexpectedly different explanations, even when applied to simple models for which we would expect qualitative coincidence. To quantify this effect, we propose a new approach to compare explanations produced by different methods.
format Preprint
id arxiv_https___arxiv_org_abs_2207_01420
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Comparing Feature Importance and Rule Extraction for Interpretability on Text Data
Lopardo, Gianluigi
Garreau, Damien
Machine Learning
Artificial Intelligence
Computation and Language
Complex machine learning algorithms are used more and more often in critical tasks involving text data, leading to the development of interpretability methods. Among local methods, two families have emerged: those computing importance scores for each feature and those extracting simple logical rules. In this paper we show that using different methods can lead to unexpectedly different explanations, even when applied to simple models for which we would expect qualitative coincidence. To quantify this effect, we propose a new approach to compare explanations produced by different methods.
title Comparing Feature Importance and Rule Extraction for Interpretability on Text Data
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2207.01420