On the Opacity of Deep Neural Networks

被引：2

作者：

Sogaard, Anders ^{[1
]}

机构：

[1] Univ Copenhagen, Ctr Philosophy AI, Copenhagen, Denmark

来源：

CANADIAN JOURNAL OF PHILOSOPHY | 2023年 / 53卷 / 03期

关键词：

deep neural networks; opacity; explainability; model size; mitigation;

D O I：

10.1017/can.2024.1

中图分类号：

B [哲学、宗教];

学科分类号：

01 ; 0101 ;

摘要：

Deep neural networks are said to be opaque, impeding the development of safe and trustworthy artificial intelligence, but where this opacity stems from is less clear. What are the sufficient properties for neural network opacity? Here, I discuss five common properties of deep neural networks and two different kinds of opacity. Which of these properties are sufficient for what type of opacity? I show how each kind of opacity stems from only one of these five properties, and then discuss to what extent the two kinds of opacity can be mitigated by explainability methods.

引用

页码：224 / 239

页数：16

共 53 条

[1] Assessing the Trustworthiness of Saliency Maps for Localizing Abnormalities in Medical Imaging [J].