Stanford's new fall course CS 312, Deep Learning Alchemy, is taught by Tatsunori Hashimoto and Suhas Kotha to build hands-on intuition for deep learning through small-model experiments.
Assessment is 7 exams worth 85% of the grade, with students allowed to use AI agents for assignments but required to show mastery on exams without them.
Assignments, quizzes and handouts are public at deep-learning-alchemy.github.io, and lecture recordings will be released through Stanford Online with a delay.
斯坦福大学 CS 312「Deep Learning Alchemy」秋季新课
@tatsu_hashimoto 和 @kothasuhas 两位老师主讲，主要面向一个常见问题：深度学习的核心知识大多是经验性的，理论解释不了，读论文也替代不了动手。
课程把两个信条作为支点：炼金术要靠手学；预测即理解：能在跑实验前准确预判结果，才算掌握。
课程地址：
https://deep-learning-alchemy.github.io/
课程安排主线
七个单元：超参调优与 scaling laws → 优化器与超参不变量（NQM、μP、Muon、权重衰减）→ 损失景观 → 训练稳定性 → 架构容量 → 数据高效的泛化（切换到 DNA 模态）→ 未公开单元。阅读清单就是实证深度学习的地图，多位作者（μP、Muon、权重衰减系列）就在团队中。
作业代码、测验、讲义全部公开，课程录像由 @StanfordOnline 发布，自学者可用 Modal 租卡完整复现。可视为 CS 336（从零构建 LM）的续作：那门课教造模型，这门课教理解模型行为。
Deep learning is often taught as a sanitized collection of facts and math. But work at the frontier of AI is messy and empirical. Our new class, Deep Learning Alchemy, aims to teach the art and science of deep learning in the hands-on spirit of Language Modeling From Scratch. Students build intuition by running experiments on small LMs, then test it through prediction quizzes: given an unseen code diff, can you predict what happens? https://deep-learning-alchemy.github.io/ My co-instructor @kothasuhas and I think there's significant demand for classes that treat students as serious learners who want to be assessed for what they’ve learned, and so we're trying a new exam-heavy approach to the class (7 exams, 85% of the grade, with the remaining 15% in-person discussion assessments). We’ve designed the class so students can use AI agents to run their experiments and do their assignment, but they then have to demonstrate that mastery on exams without access to the assignment, and any partial credit on the exam has to come through citations to the assignment, incentivizing them to know what’s in their write-up. The same assignments and quizzes are public for anyone to try (lecture recordings will be released through @StanfordOnline with some delay). We have some very fun assignments and quizzes covering both classic topics (like NQM, run-to-run stochasticity, and scaling laws) and new and timely topics like muP or hyperball. Made by some real experts like @TristanThrush and @wen_kaiyue Try some quizzes below!
Source: meng shao · x.comPublished · added here
