Authors:	Burkhardt, Sophie Brugger, Jannis Wagner, Nicole Ahmadi, Zahra Kersting, Kristian Kramer, Stefan
Title:	Rule extraction from binary neural networks with convolutional rules for model validation
Online publication date:	16-Nov-2021
Year of first publication:	2021
Language:	english
Abstract:	Classification approaches that allow to extract logical rules such as decision trees are often considered to be more interpretable than neural networks. Also, logical rules are comparatively easy to verify with any possible input. This is an important part in systems that aim to ensure correct operation of a given model. However, for high-dimensional input data such as images, the individual symbols, i.e. pixels, are not easily interpretable. Therefore, rule-based approaches are not typically used for this kind of high-dimensional data. We introduce the concept of first-order convolutional rules, which are logical rules that can be extracted using a convolutional neural network (CNN), and whose complexity depends on the size of the convolutional filter and not on the dimensionality of the input. Our approach is based on rule extraction from binary neural networks with stochastic local search. We show how to extract rules that are not necessarily short, but characteristic of the input, and easy to visualize. Our experiments show that the proposed approach is able to model the functionality of the neural network while at the same time producing interpretable logical rules. Thus, we demonstrate the potential of rule-based approaches for images which allows to combine advantages of neural networks and rule learning.
DDC:	004 Informatik 004 Data processing
Institution:	Johannes Gutenberg-Universität Mainz
Department:	FB 08 Physik, Mathematik u. Informatik
Place:	Mainz
ROR:	https://ror.org/023b0x485
DOI:	http://doi.org/10.25358/openscience-6523
Version:	Published version
Publication type:	Zeitschriftenaufsatz
License:	CC BY
Information on rights of use:	https://creativecommons.org/licenses/by/4.0/
Journal:	Frontiers in artificial intelligence 4
Pages or article number:	642263
Publisher:	Frontiers Media
Publisher place:	Lausanne
Issue date:	2021
ISSN:	2624-8212
Publisher URL:	https://doi.org/10.3389/frai.2021.642263
Publisher DOI:	10.3389/frai.2021.642263
Appears in collections:	JGU-Publikationen