Minimal explanations in ReLU-based neural network via abduction

Ariel Miki Abraham, Ari Saptawijaya, Raja O.P. Damanik

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Abduction is a well-known reasoning approach to compute plausible explanations for an observation. It has recently been employed to explain the machine learning prediction of samples from a data set by generating subset-minimal or cardinality-minimal explanations with respect to features. In this paper, we study some complexity properties of such minimal explanations in explaining predictions of neural networks. This paper also extends existing works by proposing a randomized subset-minimal procedure as a strategy to compute subset-minimal explanations. The experiment results on a number of benchmarks validate that the resulting explanations are generally smaller than subset-minimal ones. On the other hand, this strategy is not as expensive as computing cardinality-minimal explanations. It thus serves as a trade-off between the existing strategies of cardinality-and subset-minimal explanations.

Original languageEnglish
Title of host publication2020 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages61-68
Number of pages8
ISBN (Electronic)9781728192796
DOIs
Publication statusPublished - 17 Oct 2020
Event12th International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020 - Virtual, Depok, Indonesia
Duration: 17 Oct 202018 Oct 2020

Publication series

Name2020 International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020

Conference

Conference12th International Conference on Advanced Computer Science and Information Systems, ICACSIS 2020
CountryIndonesia
CityVirtual, Depok
Period17/10/2018/10/20

Fingerprint Dive into the research topics of 'Minimal explanations in ReLU-based neural network via abduction'. Together they form a unique fingerprint.

Cite this