TL;DR本研究提出了一种自我解释的 AI 的概念，它能够解释其决策的机制并提供信心水平，但由于深度神经网络的双下降现象，训练复杂真实世界数据的神经网络难以解释和外推，因此深度学习系统需要包括特定预警来保证其应用的安全性。
Abstract
The ability to explain decisions made by ai systems is highly sought after,
especially in domains where human lives are at stake such as medicine or
autonomous vehicles. While it is often possible to approximate the input-output
relations of →