Determinism and Scalability in Post-Training RL Systems

​🎙️ About the Event

​Join us for an insightful session with Ethan Su, a leading LLM Researcher and Engineer from the SGL-project and Tsinghua University PhD.

​In this talk, Ethan will present his team's advanced experiments on reinforcement learning (RL) training. He will introduce a suite of RL systems they have contributed to the open-source community, which focus on critical areas such as:

• ​Rollout optimization

• ​Cross-hardware support

• ​Scalable training while preserving determinism

​Discover the capabilities of their enterprise-grade open-source RL frameworks, slime and miles, and learn how they are pushing the boundaries of what's possible in reinforcement learning.

​👨‍💻 Speaker

​Ethan Su (@thu_yushengsu)

• ​LLMSys, LLM Researcher / Engineer

• ​SGL-project, Tsinghua University PhD

​🗓️ When & Where

• ​Date: December 6th

• ​Time: 2:00 PM - 3:00 PM KST/JST

• ​Location: Network School

​Hosted by @aerlabs.
🌐 Website: https://aerlabs.tech/
✍️ Blog: https://aerlabs.tech/blogs
📺 YouTube: https://youtube.com/@aerlabs
💼 LinkedIn: https://linkedin.com/company/aer-labs
💬 Discord: https://discord.gg/cR9Vn9zR