期刊
KNOWLEDGE-BASED SYSTEMS
卷 255, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.knosys.2022.109725
关键词
XAI; Explainable AI; Hierarchical; Middle-level; Interpretable models
资金
- European Union - FSE-REACT-EU, PON Research and Innovation 2014-2020 [DM1062/2021, 18-I-15350-2]
- Ministry of University and Research, PRIN research project BRIO - BIAS, RISK, OPACITY in AI: design, verification and development of Trustworthy AI. [2020SSKZ7R]
A central issue in eXplainable Artificial Intelligence (XAI) is to provide explanations for the behaviors of non-interpretable machine learning models. This paper proposes an XAI framework that utilizes auto-encoders to extract middle-level input features and generate explanations. Experimental results demonstrate the potential applicability of this method in image classification.
A central issue addressed by the rapidly growing research area of eXplainable Artificial Intelligence (XAI) is to provide methods to give explanations for the behaviours of Machine Learning (ML) non -interpretable models after the training. Recently, it is becoming more and more evident that new directions to create better explanations should take into account what a good explanation is to a human user. This paper suggests taking advantage of developing an XAI framework that allows producing multiple explanations for the response of image a classification system in terms of potentially different middle-level input features. To this end, we propose an XAI framework able to construct explanations in terms of input features extracted by auto-encoders. We start from the hypothesis that some auto-encoders, relying on standard data representation approaches, could extract more salient and understandable input properties, which we call here Middle-Level input Features (MLFs), for a user with respect to raw low-level features. Furthermore, extracting different types of MLFs through different type of auto-encoders, different types of explanations for the same ML system behaviour can be returned. We experimentally tested our method on two different image datasets and using three different types of MLFs. The results are encouraging. Although our novel approach was tested in the context of image classification, it can potentially be used on other data types to the extent that auto-encoders to extract humanly understandable representations can be applied. (c) 2022 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据