# An Uncertainty Principle for Linear Recurrent Neural Networks

arXiv: https://arxiv.org/abs/2502.09287 · HTML: https://mlsanity.com/paper/2502.09287/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Feb 2025.

## Built on by (most cited first, 2 of 2)

- [Selective Rotary Position Embedding](https://mlsanity.com/paper/2511.17388/) (2025)
- [Spectral Filtering for Complex Linear Dynamical Systems](https://mlsanity.com/paper/2601.22400/) (2026)

## Abstract (arXiv)

We consider linear recurrent neural networks, which have become a key building block of sequence modeling due to their ability for stable and effective long-range modeling. In this paper, we aim at characterizing this ability on a simple but core copy task, whose goal is to build a linear filter of order S that approximates the filter that looks K time steps in the past (which we refer to as the shift-K filter), where K is larger than S. Using classical signal models and quadratic cost, we fully characterize the problem by providing lower bounds of approximation, as well as explicit filters that achieve this lower bound up to constants. The optimal performance highlights an uncertainty principle: the optimal filter has to average values around the K-th time step in the past with a range~(width) that is proportional to K/S.