arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.03102cs.LGcs.AIcs.CVcs.PF
收藏

使用双因子扰动的深度学习分类器鲁棒性基准测试

Benchmarking Robustness of Deep Learning Classifiers Using Two-Factor Perturbation

Wei Dai, Daniel Berleant

AI总结:

本文提出一种新的深度学习分类器鲁棒性基准测试方法,构建了包含69张图像的双因子扰动数据集,并引入四象限统计可视化工具,发现双因子扰动图像能同时提升分类器的鲁棒性和准确率。

摘要:

本文补充了关于深度学习(DL)分类器鲁棒性基准测试的基础性研究工作。我们创新了一种新的基准测试方法来评估DL分类器的鲁棒性。同时,我们引入了一个新的四象限统计可视化工具,包括最小准确率、最大准确率、平均准确率和变异系数,用于对DL分类器的鲁棒性进行基准测试。为了测量鲁棒的DL分类器,我们创建了一个包含69个基准测试图像的综合图像集,包括一个干净集、带有单因子扰动的图像集,以及带有双因子扰动条件的图像集。在收集实验结果后,我们首次报告称,使用双因子扰动图像能够同时提升DL分类器的鲁棒性和准确率。双因子扰动包括:(1)以两种顺序应用的两种数字扰动(椒盐噪声和高斯噪声),以及(2)以两种顺序应用的一种数字扰动(椒盐噪声)和一种几何扰动(旋转)。所有源代码、相关图像集以及初步数据、图表均已在GitHub网站上分享,以支持未来的学术研究和工业项目。网络资源位于https://github.com/caperock/robustai。

英文摘要:

This paper adds to the fundamental body of work on benchmarking the robustness of deep learning (DL) classifiers. We innovate a new benchmarking methodology to evaluate robustness of DL classifiers. Also, we introduce a new four-quadrant statistical visualization tool, including minimum accuracy, maximum accuracy, mean accuracy, and coefficient of variation, for benchmarking robustness of DL classifiers. To measure robust DL classifiers, we created a comprehensive 69 benchmarking image set, including a clean set, sets with single factor perturbations, and sets with two-factor perturbation conditions. After collecting experimental results, we first report that using two-factor perturbed images improves both robustness and accuracy of DL classifiers. The two-factor perturbation includes (1) two digital perturbations (salt & pepper noise and Gaussian noise) applied in both sequences, and (2) one digital perturbation (salt & pepper noise) and a geometric perturbation (rotation) applied in both sequences. All source codes, related image sets, and preliminary data, figures are shared on a GitHub website to support future academic research and industry projects. The web resources locate at https://github.com/caperock/robustai

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。

↑