arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1905.07112cs.CRcs.AIcs.LG
收藏

对DeepSec深度学习模型安全分析平台的批评

A critique of the DeepSec Platform for Security Analysis of Deep Learning Models

Google Research(谷歌研究院)

Nicholas Carlini

摘要:

在IEEE S&P 2019上,论文《DeepSec:深度学习模型安全分析的统一平台》旨在“系统地评估现有的对抗攻击和防御方法。“虽然这篇论文的目标值得称赞,但它未能实现这些目标,并提出了根本上有缺陷和误导性的结果。我们解释了DeepSec工作中的缺陷,以及它的分析如何未能有意义地评估各种攻击和防御。具体来说,DeepSec(1)使用针对不设防模型的攻击来评估每个防御系统;(2)使用严重低估其有效性的错误实现来评估攻击和防御;(3)将每个防御系统的鲁棒性作为平均值进行评估,而不是基于对该防御系统的最有效攻击;(4)错误地执行了几次统计分析,并且未能报告方差;(5)由于这些错误,得出无效的结论,并作出笼统的概括。

英文摘要:

At IEEE S&P 2019, the paper "DeepSec: A Uniform Platform for Security Analysis of Deep Learning Model" aims to to "systematically evaluate the existing adversarial attack and defense methods." While the paper's goals are laudable, it fails to achieve them and presents results that are fundamentally flawed and misleading. We explain the flaws in the DeepSec work, along with how its analysis fails to meaningfully evaluate the various attacks and defenses. Specifically, DeepSec (1) evaluates each defense obliviously, using attacks crafted against undefended models; (2) evaluates attacks and defenses using incorrect implementations that greatly under-estimate their effectiveness; (3) evaluates the robustness of each defense as an average, not based on the most effective attack against that defense; (4) performs several statistical analyses incorrectly and fails to report variance; and, (5) as a result of these errors draws invalid conclusions and makes sweeping generalizations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑