arXiv:2410.22730cs.NEcs.LG2024-10

RNN的扩展性属性无法被自动验证,这是由理论极限决定的。

Extensional Properties of Recurrent Neural Networks

  • 从函数角度定义RNN的扩展性属性,关注其计算行为而非算法实现
  • 证明了任何非平凡的扩展性属性都是不可判定的
  • 适用于理论研究者,尤其关注RNN可验证性的方向

一个递归神经网络(RNN)的性质被称为扩展性,若它描述的是RNN所计算函数的特性,而非其算法实现本身。许多在RNN中值得关注的性质属于扩展性,例如对输入微小变化的鲁棒性或输入的良好聚类能力。给定一个RNN时,自然会问它是否具备此类性质。本文对这一普遍问题给出了否定回答:我们证明了关于RNN扩展性性质的一个类似Rice定理的结果——任何非平凡的扩展性性质都是不可判定的。

原文摘要 · Abstract (English)

A property of a recurrent neural network (RNN) is called \emph{extensional} if, loosely speaking, it is a property of the function computed by the RNN rather than a property of the RNN algorithm. Many properties of interest in RNNs are extensional, for example, robustness against small changes of input or good clustering of inputs. Given an RNN, it is natural to ask whether it has such a property. We give a negative answer to the general question about testing extensional properties of RNNs. Namely, we prove a version of Rice's theorem for RNNs: any nontrivial extensional property of RNNs is undecidable.

RNN可判定性理论分析

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。