https://www.youngju.dev/blog 천천히 올바르게. AI Researcher & DevOps Engineer Youngju's tech blog. GPU/CUDA, LLM, MLOps, Kubernetes AI workloads, distributed training, and data engineering. ko fjvbn2003@gmail.com (Youngju Kim) fjvbn2003@gmail.com (Youngju Kim) Fri, 12 Jun 2026 00:00:00 GMT https://www.youngju.dev/blog/ai/2026-06-12-imitation-vs-discovery-rich-sutton.en https://www.youngju.dev/blog/ai/2026-06-12-imitation-vs-discovery-rich-sutton.en Rich Sutton, the godfather of reinforcement learning, set the community ablaze in June 2026 by arguing that generative AI trained with supervised learning is an imitation model and therefore limited for genuinely new scientific discovery. We dissect his claim through the lens of the Bitter Lesson, the essential difference between imitation and RL, the counterarguments, and the practical lessons for agent design. Fri, 12 Jun 2026 00:00:00 GMT fjvbn2003@gmail.com (Youngju Kim) aireinforcement-learningllmresearchagentsbitter-lesson https://www.youngju.dev/blog/ai/2026-06-12-imitation-vs-discovery-rich-sutton.ja https://www.youngju.dev/blog/ai/2026-06-12-imitation-vs-discovery-rich-sutton.ja 強化学習の大御所 Rich Sutton が、教師あり学習ベースの生成 AI は模倣モデルであり新しい科学的発見には限界があると主張し、2026年6月のコミュニティを沸かせました。Bitter Lesson の文脈から彼の主張を解剖し、模倣学習と強化学習の本質的な違い、反論、そしてエージェント設計への実務的な教訓を整理します。 Fri, 12 Jun 2026 00:00:00 GMT fjvbn2003@gmail.com (Youngju Kim) aireinforcement-learningllmresearchagentsbitter-lesson https://www.youngju.dev/blog/ai/2026-06-12-imitation-vs-discovery-rich-sutton https://www.youngju.dev/blog/ai/2026-06-12-imitation-vs-discovery-rich-sutton 강화학습의 대부 Rich Sutton이 지도학습 기반 생성 AI는 모방 모델이라 새로운 과학적 발견에 한계가 있다고 주장하며 2026년 6월 커뮤니티를 달궜습니다. Bitter Lesson의 맥락에서 그의 주장을 해부하고, 모방 학습과 강화학습의 본질적 차이, 반론, 그리고 에이전트 설계에 주는 실무적 교훈을 정리합니다. Fri, 12 Jun 2026 00:00:00 GMT fjvbn2003@gmail.com (Youngju Kim) aireinforcement-learningllmresearchagentsbitter-lesson