---
id: 20260310-T0-14
title: "JAWS通过雅可比正则化稳定神经算子"
title_en: "JAWS Stabilizes Neural Operators via Jacobian Regularization"
url: https://ai.daily.yangsir.net/daily/20260310-T0-14
issue_date: 2026-03-10
publish_date: 2026-03-09T04:00:00.000Z
source_name: "arXiv cs.LG (ML)"
source_url: https://arxiv.org/abs/2603.05538
---

# JAWS通过雅可比正则化稳定神经算子

arXiv论文提出JAWS方法，通过空间自适应雅可比正则化解决神经算子长期 rollout 的不稳定问题。该技术缓解了谱爆炸现象，使连续动力学系统仿真误差降低40%。实验显示，在流体力学模拟中，JAWS比全局正则化方法收敛速度更快，适用于长时间序列预测场景。

## English Version

**JAWS Stabilizes Neural Operators via Jacobian Regularization**

arXiv paper proposes JAWS, using spatial-adaptive Jacobian regularization to solve neural operator long-term rollout instability. The tech mitigates spectral explosion, reducing continuous dynamics simulation error by 40%. Experiments show JAWS converges faster than global regularization in fluid simulations, suitable for long-term sequence prediction.

---

**来源**：[arXiv cs.LG (ML)](https://arxiv.org/abs/2603.05538)

**详情页**：https://ai.daily.yangsir.net/daily/20260310-T0-14

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*