可解释的人工智能如何影响人类的表现？对显著性地图行为后果的系统回顾

IF 3.4 3区计算机科学 Q2 COMPUTER SCIENCE, CYBERNETICS

International Journal of Human-Computer Interaction Pub Date : 2024-07-29 DOI:10.1080/10447318.2024.2381929

Romy Müller

引用次数: 0

摘要

显著性图谱可以解释深度神经网络如何对图像进行分类。但它们对人类真的有用吗？本报告对 68 项用户研究进行了系统回顾，发现虽然显著性图谱可以帮助人类更好地识别图像，但它们并不能帮助人类更好地识别图像。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

How Explainable AI Affects Human Performance: A Systematic Review of the Behavioural Consequences of Saliency Maps

Saliency maps can explain how deep neural networks classify images. But are they actually useful for humans? The present systematic review of 68 user studies found that while saliency maps can enha...

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

International Journal of Human-Computer Interaction 工程技术-计算机：控制论

CiteScore

9.00

自引率

21.30%

发文量

347

审稿时长

2 months

期刊介绍： The International Journal of Human-Computer Interaction addresses the cognitive, creative, social, health, and ergonomic aspects of interactive computing. It emphasizes the human element in relation to the systems and contexts in which humans perform, operate, network, and communicate, including mobile apps, social media, online communities, and digital accessibility. The journal publishes original articles including reviews and reappraisals of the literature, empirical studies, and quantitative and qualitative contributions to the theories and applications of HCI.