指出一篇声称证明机器学习无法实现类人智能的论文存在关键假设漏洞。
Barriers to Complexity-Theoretic Proofs that "AGI" Using Machine Learning is Impossible
- 质疑原论文关于数据分布的不合理假设
- 强调定义'类人智能'和模型归纳偏置的困难
- 适合关注AI理论边界与形式化论证的读者
一篇近期论文（van Rooij等，2024）声称在复杂性理论意义上证明了通过数据学习实现类人智能是不可行的。本文指出该证明依赖于对（输入, 输出）元组数据分布的未经证实的假设。我们简要讨论了这一假设所面临的两个根本障碍：一是必须精确界定‘类人’的含义，二是必须考虑特定机器学习系统所具有的归纳偏置，而这些偏置在分析中至关重要。另一项试图修复该证明的方法聚焦于数据子集，但面临如何定义这些子集的难题。
原文摘要 · Abstract (English)
A recent paper (van Rooij et al. 2024) claims to have proved that achieving human-like intelligence using learning from data is intractable in a complexity-theoretic sense. We point out that the proof relies on an unjustified assumption about the distribution of (input, output) tuples in the data. We briefly discuss that assumption in the context of two fundamental barriers to repairing the proof: the need to precisely define ``human-like," and the need to account for the fact that a particular machine learning system will have particular inductive biases that are key to the analysis. Another attempt to repair the proof, by focusing on subsets of the data, faces barriers in terms of defining the subsets.
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。