Skip to content
View original post on X: meng shao
Published
44/100

Stanford's CS329A course on building self-improving AI agents is on YouTube

AIStanford has published all nine lectures of CS329A Self-Improving AI Agents on YouTube, taught by two instructors who have worked on PaLM, Gemini, and Claude. The course centers on a loop of sampling, verification, filtering, and training to produce a stronger model.
meng shaoVerified on X
@shao__meng
View on X

斯坦福大学课程 CS329A Self-Improving AI Agents

@achowdhery @Azaliamirh 两位老师主讲,以 "一线经验 + 论文精读" 为骨架,以 "造一个能自我改进的智能体" 为课程目标,Youtube 已收录全部 9 节课程。

还没学过这个课程的朋友,可以先看看我做的这个演示视频,强烈推荐大家系统学习原课程视频,虽然花一些时间,但沉下心来系统学习思考,很有收获!

另外也欢迎朋友们对这个演示视频的制作多提宝贵意见,最近在磨一套演示视频的 Skills,想用于课程、开源项目、产品发布、文章等的快速表达演示。等我自己磨满意了会发布出来。

课程主页:http://cs329a.stanford.edu
课程视频:https://www.youtube.com/playlist?list=PLangBM27OtEA

meng shao@shao__meng
斯坦福大学把 CS329A Self-Improving AI Agents 视频都公布到 Youtube 了。 两位主讲老师做过 PaLM、Gemini、Claude。9 讲讲透一个循环: 采样 → 验证 → 筛选 → 训练 → 更强的模型 想知道为什么 coding agent 最新普及,写作却不行? 完整解读 👇 https://x.com/i/article/2096524395980468224
View quoted post on X
Source details

meng shao · x.comPublished

Original title: 斯坦福大学课程 CS329A Self-Improving AI Agents