In this ALIGN webinar, we invited the AI researcher Shota Imai, who has done AI research at the Matsuo Laboratory of the University of Tokyo and was then preparing for a new venture. His research covers reinforcement learning, multi-agent systems, generative AI, large language models and game AI, and in January 2024 he published a book on how generative AI will change the world, Seisei AI de sekai wa kō kawaru (SB Shinsho).
He explained alignment techniques for large language models, centered on reinforcement learning from human feedback (RLHF). The webinar was held in Japanese.