Learn Generative AI, Prompt Engineering, and LLMs for Free
The Fastest Way to Become a Backend Developer Online
Overview
AI, Data Science & Cloud Certificates from Google, IBM & Meta — 40% Off
One plan covers every Professional Certificate on Coursera. 40% off Coursera Plus Annual.
Unlock All Certificates
Explore a groundbreaking 26-minute video presentation on the revolutionary UR2 framework that unifies Retrieval-Augmented Generation (RAG) and reasoning through reinforcement learning. Discover how reinforcement learning represents the future evolution of RAG systems, moving beyond traditional approaches to create self-learning AI agents capable of sophisticated reasoning. Learn about the innovative research from Tsinghua University's Department of Computer Science & Technology and Institute for AI, in collaboration with Hebei University of Economics and Business, that introduces RAG 3.0 methodology. Understand the technical foundations of the UR2 (Unify RAG and Reasoning through Reinforcement Learning) framework developed by researchers Weitao Li, Boran Xiang, Xiaolong Wang, Zhinan Gou, Weizhi Ma, and Yang Liu. Gain insights into how this cutting-edge approach transforms traditional RAG systems by incorporating reinforcement learning mechanisms that enable AI agents to continuously improve their reasoning capabilities through self-learning processes, representing a significant advancement in artificial intelligence and machine learning applications.
Syllabus
RAG 3.0 in RL: Self-Learning AI Agent Reasoning (Tsinghua)
Taught by
Discover AI