An Assurance Case Pattern for the Interpretability of Machine Learning in Safety-Critical Systems

An Assurance Case Pattern for the Interpretability of Machine Learning in Safety-Critical Systems
复制标题

安全关键系统中机器学习可解释性的保证案例模式

DOI:
--
复制
发表时间:
2020
期刊:
SAFECOMP Workshops
影响因子:
--
通讯作者:
I. Habli
I. Habli
中科院分区:
--
文献类型:
--
作者:
Francis Rhys Ward;I. Habli

文献摘要

参考文献

被引文献

相似文献

机器学习(ML)有可能在安全关键应用中得到广泛应用。因此,我们必须对基于ML的功能的安全行为有足够的信心。一个关键的考虑因素是所使用的ML是否可解释。在本文中,我们提出了一种论元模式,即可重用结构,它可以用来在更广泛的保证情况下证明ML的充分可解释性。该模式可用于评估是否在正确的上下文(时间、背景和受众)中使用了正确的可解释性方法和格式。这一论证结构为开发和评估在安全关键领域对ML的可解释性的重点要求提供了基础。
Machine Learning (ML) has the potential to become widespread in safety-critical applications. It is therefore important that we have sufficient confidence in the safe behaviour of the ML-based functionality. One key consideration is whether the ML being used is interpretable. In this paper, we present an argument pattern, i.e. reusable structure, that can be used for justifying the sufficient interpretability of ML within a wider assurance case. The pattern can be used to assess whether the right interpretability method and format are used in the right context (time, setting and audience). This argument structure provides a basis for developing and assessing focused requirements for the interpretability of ML in safety-critical domains.
DOI: 10.1016/j.artint.2019.103201
发表时间: 2020-02-01
影响因子: 14.4
作者:
Burton, Simon;Habli, Ibrahim;Porter, Zoe
通讯作者: Porter, Zoe
DOI: --
发表时间: 2018-11
期刊: ArXiv
影响因子: --
作者:
C. Rudin
通讯作者: C. Rudin