[译文] 苦涩的教训 英文原文:The Bitter Lesson 理查德·萨顿(Richard S. Sutton,现代强化学习之父,现任加拿大阿尔伯塔大学教授) 2019年3月13日 过去70年人工智能研究领域最重要的一堂课是,只有通用计算方法最终是最有效的,而且优势巨大。根本原因是摩尔定律,更确切地说是,每个计算单元的成本持续呈指数下降。大多数...
one approach or the other. And thehuman-knowledgeapproach tends to complicate methods in ways that make them less suited to taking advantage of general methods leveraging computation. There were many examples of AI researchers' belated learning of this bitter lesson, and it is instructive to review...
原文链接:http://www.incompleteideas.net/IncIdeas/BitterLesson.html 因此,作者 @SIY.Z 认为某种意义上,强大算力加持的通用 AI 算法才是 AGI 路径的王道和 AI 技术真正进步的方向。有了大模型、大算力和大数据,The bitter lesson 构成了 AGI 的必要条件。再加上 Scaling Law 这一充分条件,通过算法使大模型、...
人工智能专家 Rich Sutton 2019年曾有雄文"痛苦的教训" (The bitter lesson),总结过去七十年来人工智能研究的教训,其核心观点是: 算力的大规模进步会碾压各种局部算法的改进,但研究者往往会假设算力不会改变太多,用自己主观经验和局部知识去改进 AI, 短期内会收获一些进步,研究者会有成就感,但这些改进一般最终遭遇...
one approach or the other. And the human-knowledge approach tends to complicate methods in ways that make them less suited to taking advantage of general methods leveraging computation. There were many examples of AI researchers' belated learning of this bitter lesson, and it is instructive to ...
特别是当今ChatGPT 大模型的横空出世,让人们试图重新审视人工智能的规模优势的这个发展现实。实际上,Rich Sutton 教授 --- 现代人工智能领域强化学习 (Reinforcement Learning) 鼻祖 --- 在2019年的“The Bitter Lesson (惨痛的教训)”一文中就对深度学习应用的“大力...
量子位 出品 | 公众号 QbitAI 70年来， 人们在AI领域“一直连续犯着同样的错误”。这是“强化学习之父”理查德·萨顿（Richard S. Sutton）为同行后辈们敲响的警钟。他在博客上发表最新文章《苦涩的教训》(The Bitter Lesson)，总结了AI发展史上的怪圈：人类不断试图把自己的知识和思维方式植入到AI之中，比如...
we have to understand the appeal of these mistakes. We have to learn the bitter lesson that building in how we think we think does not work in the long run. The bitter lesson is based on the historical observations that 1) AI researchers have often tried to build knowledge into their age...
In a popular blog post titled “The Bitter Lesson,” Richard Sutton argues that AI’s progress has resulted from cheaper computation, not human design decisions based on problem-specific information.
文章链接the bitter lesson 今天又读了一遍Richard Sutton的这篇博客。在这篇博客中,Sutton总结了AI研究的近几十年取得的进步的原因和教训。 文章中称AI取得显著的进步靠的不是依赖人的领域知识(例如象棋、围棋、语音识别),而是靠算力、搜索和学习(Search and Learning)。试图将人的领域知识解决问题短期内会取得一定...