一句话：系统探讨深度学习软件中对抗样本缺陷与模型不确定性之间的关系，为理解此类缺陷提供新视角。
过去十年，深度学习在众多工业任务中取得成功，但当前深度学习软件仍存在质量问题，在安全与安保关键场景中尤其令人担忧。对抗样本是一类典型且重要的缺陷，会使深度学习软件做出错误决策，其可能来自恶意攻击，也可能来自输入传感器感知到的物理世界噪声，从而阻碍工业部署。深度学习决策固有的不确定性，可能是其错误行为的根本原因。尽管已有测试、对抗攻击与防御技术，仍缺乏系统研究来揭示对抗样本与深度学习不确定性之间的关系。本文旨在从不确定性视角系统刻画深度学习软件的对抗性缺陷，为后续检测、定位与修复提供基础。摘要未给出具体方法与实验结果。
Over the past decade, deep learning (DL) has been successfully applied to many industrial domain-specific tasks. However, the current state-of-the-art DL software still suffers from quality issues, which raises great concern especially in the context of safety- and security-critical scenarios. Adversarial examples (AEs) represent a typical and important type of defects needed to be urgently addressed, on which a DL software makes incorrect decisions. Such defects occur through either intentional attack or physical-world noise perceived by input sensors, potentially hindering further industry deployment. The intrinsic uncertainty nature of deep learning decisions can be a fundamental reason for its incorrect behavior. Although some testing, adversarial attack and defense techniques have been recently proposed, it still lacks a systematic study to uncover the relationship between AEs and DL uncertainty.
DOI 原文 ·
@inproceedings{paperbot5607,
title = {Towards characterizing adversarial defects of deep learning software from the lens of uncertainty},
author = {Xiyue Zhang and Xiaofei Xie and Lei Ma and Xiaoning Du and Qiang Hu and Yang Liu and Jianjun Zhao and Meng Sun},
booktitle = {ICSE},
year = {2020},
doi = {10.1145/3377811.3380368}
}