-
Notifications
You must be signed in to change notification settings - Fork 77
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
docs: 补充 AI 伦理安全、多媒体安全课程内容,调整部分阅读书目推荐 (#187)
* 补充了 AI 伦理安全、多媒体安全两门课程, 并调整了少部分阅读书目推荐。 * Update 人工智能伦理与安全 * minor changes * minor changes
- Loading branch information
1 parent
986b674
commit 5cc6211
Showing
11 changed files
with
170 additions
and
33 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,52 @@ | ||
# 多媒体安全 | ||
<div class="badges"> | ||
<span class="badge is-badge">IS 专业选修</span> | ||
</div> | ||
|
||
|
||
## 课程学习内容 | ||
|
||
本课程旧名 "信息隐藏与数字水印", 是黄劲老师开设的课程。主要授课内容基本可以概括为 "水印" 和 "隐写" 两个部分。课程内容较多, 且十分理论, 包括 | ||
|
||
+ 水印和隐写的介绍和性质 | ||
+ 水印: | ||
+ 水印模型 | ||
+ 多比特信息嵌入 | ||
+ ECC(校验编码) | ||
+ 应用原图作为辅助增强 | ||
+ Dirty-paper code, Lattice Code | ||
+ 鲁棒水印 | ||
+ Content Authentication, 也就是 erasable 水印 | ||
+ 隐写 | ||
+ 经典算法: LSB对DCT系数(JSteg),OutGuess,F3,F4,F5(矩阵编码)以及混淆方法(random walk); | ||
+ Wet Paper, 对隐写的检测与攻击 | ||
+ 使用 Sample pairs 进行隐写分析 | ||
+ 电子取证、利用指纹信息进行身份认证等。 | ||
老师会在最后一节课进行总复习。 | ||
|
||
## 任课教师 | ||
|
||
仅有黄劲老师一位。黄老师授课很有个性, 思维跳脱、富有热情, 很喜欢跟学生交流。但是本课程内容较多也较杂, 可能稍难跟上。不过课余时间老师比较亲和, 有任何问题都会耐心解答。 | ||
|
||
## 课程教材 | ||
|
||
Cox, I., Miller, M., Bloom, J., Fridrich, J., & Kalker, T. (2007). *Digital watermarking and steganography*. Morgan kaufmann. 一本十分厚的书, 本课程内容除了取证部分外基本都是基于课本展开。虽然此书叙述<del>啰嗦</del>细致, 但也有未能讲清楚的地方。 | ||
|
||
## 分数构成 | ||
一般来说分数构成为: lab 40\%, 期末 40\%, 课堂表现 20\%, 可能有稍许浮动。 | ||
+ lab 为 4~5 次, 基本包括 | ||
+ E_BLIND/D_LC 水印系统实现 | ||
+ E_SIMPLE_8/D_SIMPLE_8 多位信息水印系统实现 | ||
+ E_BLK_8/D_BLK_8 w/ Hamming Code 或 Trellis Code 水印系统实现 | ||
+ F4, F5 隐写算法实现 | ||
|
||
lab 不限定编程语言。代码量约在数百行, 虽不算难但也对编程能力稍有要求。 | ||
+ 期末开卷, 包括少量选择、判断以及 8 道简答, 主要侧重课程内容的理解。要求英文作答。 | ||
+ 课堂表现分为两个部分: | ||
+ 老师不会点名, 但会随堂 quiz 9 次之多 :-) | ||
+ 每次课前会让两人一组进行一个简短的展示, 10 分钟左右; Topic 包括书中未讲到的部分, 也包括 lab 实现等。 | ||
|
||
## 参考资料 | ||
|
||
+ https://www.cc98.org/topic/5362725 | ||
+ https://github.com/QinJiuJiu/Information-hiding-and-digital-watermarking |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,3 +1,52 @@ | ||
--- | ||
abbrs: | ||
- AI 伦理安全 | ||
--- | ||
# 人工智能伦理与安全 | ||
<div class="badges"> | ||
<span class="badge is-badge">AI 专业必修</span> | ||
</div> | ||
|
||
\#TODO: 目前还没有内容,待完善。 | ||
|
||
|
||
## 学习内容 | ||
|
||
+ 伦理部分(摘自潘恩荣老师课件) | ||
+ 第1次 人工智能与伦理 | ||
+ 第2次 人工智能伦理谱系 | ||
+ 第3次 人工智能如何以人为本? | ||
+ 第4次 人工智能伦理研究与设计方法 | ||
+ 第5次 应用场景与伦理:自动驾驶专题 | ||
+ 第6次 数据与伦理:隐私保护专题 | ||
+ 第7次 算法与伦理:“人工智能+”专题 | ||
+ 第8次 算力与伦理:机器换人专题 | ||
+ 安全部分 | ||
+ 包括概论、介绍对抗样本攻击与 AI 隐私相关攻击 | ||
|
||
安全部分内容与 IS 专业[人工智能安全](https://zju-turing.github.io/TuringCourses/major_mandatory/ai_security/)课程完全相同。<del>估计学院没想到会有人把这俩门课都选了……</del> | ||
|
||
## 授课老师 | ||
|
||
伦理部分由马院的潘恩荣老师授课。根据查老师以及笔者本人体验, 大家对他评价普遍较低。 | ||
<!-- 别听 --> | ||
|
||
安全部分由计院的杨子祺老师授课。杨老师还是很不错的, 授课比较清晰, 他本人也是从事 AI 安全方面研究。 | ||
|
||
## 分数构成 | ||
课堂表现占 20\% , 作业占 80\%。作业仅包括一篇伦理论文和一篇安全论文, 由小组组队完成。2-5人自由组队。 | ||
课堂表现包括 | ||
+ 潘老师上课点名, | ||
+ 关于 AI 论题的一次辩论。两个小组组成辩论组, 自行寻找辩题。 | ||
+ 《人工智能伦理》网课, 以看完网课占一半考勤分数。 | ||
|
||
2023-24 春夏选题如下: | ||
+ 伦理方面 | ||
+ 根据辩论内容, 提出一个伦理问题的应对方案, 可用课程中的理论与方法进行分析, 从技术角度提出应对伦理问题的思路。 | ||
+ 安全方面 | ||
+ 完成一份对抗样本攻防或者人工智能数据隐私保护的调研报告。 | ||
+ CV、NLP、语音、LLM 方面的对抗样本生成、检测、防御等均可。 参考主题包括鲁棒性研究、对抗性训练、大模型越狱攻击等。 | ||
+ 数据窃取、隐私防御等隐私保护方向均可。参考主题: ML 模型隐私攻防、大模型数据隐私推断。 | ||
至于报告内容、结构, 会给出具体的模板。 | ||
|
||
## 学习建议 | ||
本课程有一种 "思政课和专业课" 强行拼成一节课的割裂感, 给笔者的体验较差, 尤其伦理部分。对杨老师安全部分若有感兴趣的同学不妨一听。建议水过。 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.