Stanford CME295 Transformers & LLMs | Autumn 2026 | Lecture 4 - Reinforcement Learning with LLMs

Name: Stanford CME295 Transformers & LLMs | Autumn 2026 | Lecture 4 - Reinforcement Learning with LLMs
Uploaded: 2026-10-19T00:00:00.000Z
Description: CME 295’s fourth lecture derives how reinforcement learning turns an already pretrained and supervised-fine-tuned LLM into a policy that maximizes rewards, then shows why raw policy gradients fail and how PPO, GRPO, a...

0:00 · Watch on YouTube

CME 295’s fourth lecture derives how reinforcement learning turns an already pretrained and supervised-fine-tuned LLM into a policy that maximizes rewards, then shows why raw policy gradients fail and how PPO, GRPO, a...

Read the full summary on retube