Generative AI Meets Reinforcement Learning: Generative AI Meets Reinforcement Learning

Amy Zhang, Benjamin Eysenbach

International Conference on Machine Learning 2025 · Tutorial

This comprehensive tutorial, "Generative AI Meets Reinforcement Learning," delivered by Amy Zhang and Benjamin Eysenbach at ICML 2025, explores the profound and often overlooked synergies between **generative AI (GenAI)** and **reinforcement learning (RL)**. The speakers argue that while superficially distinct, these two fields offer crucial tools and perspectives that can drive significant progress in each other. Benjamin Eysenbach opens with a historical analogy to the "Mechanical Turk," highlighting that current generative models, like early chess AIs, are often no smarter than the human data used to train them, merely mimicking existing patterns. The central thesis is that to advance beyond mere imitation, both fields must leverage each other's strengths.

AI review

A competent and well-organized tutorial by Zhang and Eysenbach that surveys the connections between generative modeling and reinforcement learning at ICML 2025. The conceptual framing is genuinely useful — the probabilistic reinterpretation of the RL objective via ELBO, the treatment of denoising as an MDP, and the occupancy-measure perspective on self-supervised RL are each legitimate theoretical threads worth weaving together. The presentation hits the right literature (FB representations, DIAYN, Diffusion-DICE, Decision Transformer) and constructs a coherent narrative around them. That…