Tingyu Zhang

Thoughts on reinforcement learning, math, and engineering

a post with plotly.js

Trial for uploading PDF onto Blog.

1 min read · March 08, 2026

2026 · formatting charts · sample-posts
Understanding GRPO from numerical optimization

This is the pdf I used for illustrating what is GRPO from numerical optimization view.

1 min read · March 08, 2026

2026 · Optimization Numerical_Analysis DeepSeek_R1 · AI