Group Relative Policy Optimization (GRPO)

Concept

A novel training method developed by DeepSeek to improve model efficiency and accuracy, also used in DeepSeek R1.

Mentioned in 1 video

Videos Mentioning Group Relative Policy Optimization (GRPO)

"OpenAI is Not God” - The DeepSeek Documentary on Liang Wenfeng, R1 and What's Next

"OpenAI is Not God” - The DeepSeek Documentary on Liang Wenfeng, R1 and What's Next

AI Explained

A novel training method developed by DeepSeek to improve model efficiency and accuracy, also used in DeepSeek R1.