Adversarial-Deep-Learning / Adversarial-Deep-Learning/code-soup

Non targeted One Pixel Attack on random 500 samples

未關閉
#78 2 則留言 0 個 reaction 已指派 0 人 在 GitHub 檢視
good first issue Priority:Medium Tutorials
主要語言
Jupyter Notebook
星號
17
分支
17
PR 合併指標
30 天內沒有已合併 PR

描述

This will be one of the tutorials for One Pixel Attack using VGG16 and CIFAR10 (open to discussion if difficult). Try to include the visualizations in original paper. [Link](https://arxiv.org/pdf/1710.08864.pdf)

The following is the minimum requirement:-

- [ ] attack on 500 random samples
- [ ] Attack using 1/3/5 Pixel(s)
- [ ] Accuracy and Confidence analysis
- [ ] Pre and Post attack Image visualizations

貢獻指南

開啟貢獻指南

評估

這個 Issue 還沒有評估資料。

把新 issue 寄到你的電子郵件信箱

精選適合新手參與的 GitHub issue 摘要。