[COURSE] Add MIT 6.S184 (#772)

* Add course

* add nav

docs/深度生成模型/MIT6.S184.en.md

@@ -0,0 +1,20 @@
+# MIT6.S184: Generative AI with Stochastic Differential Equations
+
+## Course Introduction
+
+- University: MIT  
+- Prerequisites: Basic understanding of deep learning, and be comfortable with calculus and linear algebra  
+- Programming Language: Python (with PyTorch)  
+- Course Difficulty: 🌟🌟🌟🌟  
+- Estimated Study Hours: 20  
+
+This course is an introductory diffusion model course offered during MIT's IAP term by MIT CSAIL. Taught by MIT students Peter Holderrieth and Ezra Erives, the course provides a clear and accessible explanation of the mathematical foundations of diffusion and flow-matching models from the perspective of differential equations. It also includes hands-on labs where students build diffusion models from scratch, concluding with lectures on applications in cutting-edge areas such as molecular design and robotics.
+
+The accompanying lecture notes are exceptionally well-written and highly recommended for in-depth reading.
+
+## Course Resources
+
+- Course Website: https://diffusion.csail.mit.edu/  
+- Course Videos: See course website  
+- Course Textbook: [An Introduction to Flow Matching and Diffusion Models](https://arxiv.org/abs/2506.02070)  
+- Course Assignments: Three labs, see course website for details  

docs/深度生成模型/MIT6.S184.md

@@ -0,0 +1,20 @@
+# MIT6.S184: Generative AI with Stochastic Differential Equations
+
+## 课程简介
+
+- 所属大学：MIT
+- 先修要求：Basic understanding of deep learning, and be comfortable with calculus and linear algebra
+- 编程语言：Python (with PyTorch)
+- 课程难度：🌟🌟🌟🌟
+- 预计学时：20
+
+这门课程是由 MIT CSAIL 的 IAP 小学期开办的扩散模型入门课程。该课程由 MIT 学生 Peter Holderrieth 和 Ezra Erives 主讲，从微分方程的视角深入浅出地讲解了扩散模型和流匹配模型的数学理论基础，并且配以实践让学生从零构建扩散模型，最后通过讲座介绍其在分子设计和机器人学等前沿技术中的应用。
+
+课程配套的教材笔记写得非常好，推荐仔细阅读。
+
+## 课程资源
+
+- 课程网站：https://diffusion.csail.mit.edu/
+- 课程视频：参见课程网站
+- 课程教材：[An Introduction to Flow Matching and Diffusion Models](https://arxiv.org/abs/2506.02070)
+- 课程作业：三个实验，具体参见课程网站

docs/深度生成模型/roadmap.en.md

@@ -6,6 +6,8 @@ In fact, LLMs are just one branch of deep generative models. Other types such as
 
 Recommended courses for learning:
 
+- [MIT 6.S184: Generative AI with Stochastic Differential Equations](./MIT6.S184.md): An introductory GenAI course offered during MIT's IAP term. It explains the mathematical foundations behind Flow Matching and Diffusion Models from the perspective of differential equations, accompanied by simple hands-on labs to help students grasp the concepts through practice. Ideal for those interested in the underlying mathematical principles.
+
 - [MIT 6.S978: Deep Generative Models](https://mit-6s978.github.io/schedule.html): Taught by MIT’s rising star Prof. Kaiming He, this course covers fundamental theories and cutting-edge papers related to various generative models. The assignments include well-prepared scaffold code. While not overly difficult, they help deepen understanding and provide a quick, comprehensive view of the field.
 
 - [UCB CS294-158-SP24: Deep Unsupervised Learning](https://sites.google.com/view/berkeley-cs294-158-sp24/home): Taught by reinforcement learning giant Pieter Abbeel. Compared to the MIT course, this one is more comprehensive and includes lecture videos and slides. The homework only provides test code, so students must implement model architecture and training code themselves. Though demanding, it’s ideal for those who want hands-on experience in training models. As is well known, there are many practical tricks in deep learning, and the devil is often in the details. Nothing teaches those details better than training a model yourself.

docs/深度生成模型/roadmap.md

@@ -5,6 +5,8 @@
 其实，大语言模型只是深度生成模型的一个分支，而其他生成模型例如 VAE，GAN，Diffusion Model，Flow 等等，都还在“生成”这一领域占有重要地位，所谓的 AIGC，就是泛指这一类技术。
 推荐学习下列课程：
 
+- [MIT 6.S184: Generative AI with Stochastic Differential Equations](./MIT6.S184.md): MIT IAP 小学期的 GenAI 入门课程，主要通过微分方程的视角讲解了 Flow Matching 和 Diffusion Model 背后的数学原理，并且配有简单的小实验让学生在实践中理解，适合对底层数学原理感兴趣的同学入门。
+
 - [MIT 6.S978: Deep Generative Models](https://mit-6s978.github.io/schedule.html): MIT 新晋明星教授何恺明亲授，涵盖了各种生成模型的基础理论和相关前沿论文，几次作业都有丰富的脚手架代码，难度不高但能加深理解，能对这个领域有个快速全貌了解。
 
 - [UCB CS294-158-SP24: Deep Unsupervised Learning](https://sites.google.com/view/berkeley-cs294-158-sp24/home): 强化学习领域的顶级巨佬 Pieter Abbeel 主讲，相比 MIT 的课程内容更加丰富全面，并且有配套课程视频和 Slides。此外课后作业只有测试代码，需要学生自主编写模型架构定义和训练代码，虽然硬核但很适合有志于炼丹的同学练手。众所周知，深度学习理论实践中存在着很多经验技巧，魔鬼往往存在于细节里。没有什么比自己上手训一个模型更能掌握这些细节了。