← 研踪 PaperTrailarXiv:2611.24394 · v1
Temporal Rank in Recurrent Neural Networks
首次提交 2026-11-18 21:28
版本更新 2026-11-18 21:28
雷达观测 2026-11-19 15:04
摘要
Recurrent Neural Networks (RNNs) are a natural framework for temporal modelling, yet are often dismissed due to their sequential, non-parallelisable computation and difficulty in capturing long-range dependencies, typically attributed to vanishing and exploding gradients. We identify a distinct and under-explored limitation: an information bottleneck induced by rank-constrained hidden state dynamics, which limits the number of independent input directions that can be preserved over time, as reflected in the rank of the input-to-state mapping. We show that this constraint restricts temporal capacity independently of gradient stability. To address this, we introduce structured high-dimensional state dynamics that preserve temporal separability and prevent saturation of the input-to-state Jacobian over time. On controlled sequence tasks, the resulting model exhibits rapid learning despite remaining strictly within a standard RNN formulation. Our results demonstrate that RNN performance is governed by the evolution and preservation of temporal rank, and suggest that mitigating gradient pathologies alone is insufficient without explicitly maintaining temporal capacity.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-18 21:28 · 最新观测版本
Temporal Rank in Recurrent Neural Networks
Recurrent Neural Networks (RNNs) are a natural framework for temporal modelling, yet are often dismissed due to their sequential, non-parallelisable computation and difficulty in capturing long-range dependencies, typically attributed to vanishing and exploding gradients. We identify a distinct and under-explored limitation: an information bottleneck induced by rank-constrained hidden state dynamics, which limits the number of independent input directions that can be preserved over time, as reflected in the rank of the input-to-state mapping. We show that this constraint restricts temporal capacity independently of gradient stability. To address this, we introduce structured high-dimensional state dynamics that preserve temporal separability and prevent saturation of the input-to-state Jacobian over time. On controlled sequence tasks, the resulting model exhibits rapid learning despite remaining strictly within a standard RNN formulation. Our results demonstrate that RNN performance is governed by the evolution and preservation of temporal rank, and suggest that mitigating gradient pathologies alone is insufficient without explicitly maintaining temporal capacity.
查看此版本解读 →查看此版本原文 ↗