⚡ CrackLab
advanced~12h

GRPO: Group Relative Policy Optimization

Master DeepSeek's reinforcement learning algorithm, group relative advantages, and training loops.

4
Subtopics
1
Exercises
1
Projects
1
Quiz Qs
1
Flashcards
▶📚 Prerequisites(2)

🎓 Learning objectives

  • •Explain how GRPO replaces the Critic model with group statistics
  • •Compute relative advantages from a group of rewards
  • •Design verification-based reward functions

GRPO: Group Relative Policy Optimization is a Pro topic

Sign in, then upgrade to Pro or Power to unlock this topic and the full AI Engineering curriculum.

📂 Subtopics

1

Why GRPO: The SFT vs. RFT vs. RLHF Decision Tree

2

How GRPO Works: Groups, Rewards, and the GRPO Loss

3

Reward Signal Design: 4 Deterministic Reward Functions

4

Hands-On: Fine-Tuning a Reasoning Model with Unsloth and HuggingFace TRL

Related concepts

reinforcement-learningllm-training

Next to learn

finetuning-peft →agent-patterns →
HomeGo to Dashboard