Stanford AI 课程地图 EN GitHub ↗
← 返回课程地图
Stanford · 2026—27 CS 224R 基础核心 —

延伸课程 · 基础核心

深度强化学习

Deep Reinforcement Learning

从模仿学习到 RLHF,偏深度与机器人/LLM 应用的强化学习。

3 学分 2026–27 未排课 Letter or Credit/No Credit —
深度强化学习

课程定位

从模仿学习到 RLHF,偏深度与机器人/LLM 应用的强化学习。 本页保留官方档案与入口,具体内容与考核以当季官网为准。

主要内容

本页为延伸课程速览,主要内容见下方官方描述。

课程大纲 / 课件

#主题 / 讲者
01Course Intro + Start of MDPs & Imitation
02Imitation Learning
03Policy Gradients
04Actor-Critic Methods
05Off-Policy Actor-Critic Methods
06Q-learning
07Offline RL
08Reward Learning
09RL for LLMs: Preference Optimization (Archit Sharma)
10RL for LLMs: Reasoning (Noam Brown)
11Model-Based RL
12Multi-Task and Goal-Conditioned RL
13Meta-RL
14Hierarchical RL and IL
15RL for Robots: Sim-to-Real Transfer (Guanya Shi)
16RL for Robots: RL for VLAs
17Frontiers

Spring 2026 timeline;17 讲 slides 已直接链接官方 PDF。

先修

见官方描述(延伸课程暂未单独整理先修)。

考核与工作量

以当季官网为准(延伸课程暂未单独整理考核)。

适合谁

对「深度强化学习」方向感兴趣的人。

学习建议

先读官方描述和课程官网,再判断是否适合自己。

相关课程

官方课程描述(英文原文)

Humans, animals, and robots faced with the world must make decisions and take actions in the world. Moreover, the decisions they choose affect the world they exist in - and those outcomes must be taken into account. This course is about algorithms for deep reinforcement learning - methods for learning behavior from experience, with a focus on practical algorithms that use deep neural networks to learn behavior from high-dimensional observations. Topics will include methods for learning from demonstrations, both model-based and model-free deep RL methods, methods for learning from offline datasets, and more advanced techniques for learning multiple tasks such as goal-conditioned RL, meta-RL, and unsupervised skill discovery. These methods will be instantiated with examples from domains with high-dimensional state and action spaces, such as robotics, visual navigation, and control. This course is complementary to CS234, which neither being a pre-requisite for the other. In comparison to CS234, this course will have a more applied and deep learning focus and an emphasis on use-cases in robotics and motor control.