资讯

This new framework lets LLM agents learn from experience, no fine-tuning required

Instead of retraining the LLM, the agent consults a dynamic store of past outcomes to make smarter decisions for new tasks.

TreePO Technology Innovation in AI Training: Making Reinforcement Learning Smarter and More ...

Released in August 2025, this research introduces a new method called TreePO (Tree-structured Policy Optimization), aimed at ...
Princeton University16 年

ICML2009 tutorial: The Neuroscience of Reinforcement Learning

This tutorial will present the current state of the study of neural reinforcement learning, with an emphasis on both what it teaches us about the brain, and what it teaches us about reinforcement ...

14B Surpasses 671B! Microsoft's rStar2-Agent Mathematical Reasoning Exceeds DeepSeek-R1 ...

Breakthroughs in Agentic Reinforcement Learning The success of rStar2-Agent can be attributed to three major innovations in ...

CoreWeave to acquire OpenPipe, a Seattle-area startup that uses reinforcement learning to ...

CoreWeave said it will acquire OpenPipe, a Bellevue, Wash.-based startup that helps developers train AI agents using ...

CoreWeave acquires agent-training startup OpenPipe

CoreWeave hopes the YC-backed startup will help it expand up the stack and cash in on enterprises developing AI agents.
Forbes5 年

Reinforcement Learning: The Next Big Thing For AI (Artificial ...

Reinforcement learning has even been shown to be effective when finding better solutions for tax policies and equality, as seen with Saleforce.com’s AI Economist.