厘清AGI本质：用适应性定义智能，融合搜索与近似构建通用系统。
What the F*ck Is Artificial General Intelligence?
- 以适应性为核心定义智能，将AGI视为人工科学家。
- 指出当前主流依赖规模扩展的近似方法，但效率瓶颈已现。
- 提出三类元方法：扩规模、求简洁、弱约束，适合长期研究者参考。
人工智能通识智能（AGI）是成熟的研究领域，但其概念常被过度炒作而模糊。梅琳娜·米切尔认为，唯有通过长期科学探索才能澄清争议。本文简明而富有启发性地概述了AGI，对比不同智能定义，最终以适应性为基准，将AGI定位为人工科学家。借鉴萨顿的‘苦教训’，强调搜索与近似是构建自适应系统的两大基石。比较了o3、AlphaGo、AERA、NARS、Hyperon等架构的优劣与融合可能。进一步提出三类元方法：基于苦教训的规模最大化（scale-maxing）、基于奥卡姆剃刀的简约最大化（simp-maxing）、基于本内特剃刀的约束弱化（w-maxing）。涵盖AIXI、自由能原理及语言模型的‘膨胀化’等实例。尽管当前以规模扩展的近似主导，但未来AGI将是多种工具与元策略的融合。硬件进步推动了‘膨胀化’，如今瓶颈转向样本与能源效率。
原文摘要 · Abstract (English)
Artificial general intelligence (AGI) is an established field of research. Yet some have questioned if the term still has meaning. AGI has been subject to so much hype and speculation it has become something of a Rorschach test. Melanie Mitchell argues the debate will only be settled through long term, scientific investigation. To that end here is a short, accessible and provocative overview of AGI. I compare definitions of intelligence, settling on intelligence in terms of adaptation and AGI as an artificial scientist. Taking my cue from Sutton's Bitter Lesson I describe two foundational tools used to build adaptive systems: search and approximation. I compare pros, cons, hybrids and architectures like o3, AlphaGo, AERA, NARS and Hyperon. I then discuss overall meta-approaches to making systems behave more intelligently. I divide them into scale-maxing, simp-maxing, w-maxing based on the Bitter Lesson, Ockham's and Bennett's Razors. These maximise resources, simplicity of form, and the weakness of constraints on functionality. I discuss examples including AIXI, the free energy principle and The Embiggening of language models. I conclude that though scale-maxed approximation dominates, AGI will be a fusion of tools and meta-approaches. The Embiggening was enabled by improvements in hardware. Now the bottlenecks are sample and energy efficiency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。