研究人员开发了傅里叶神经算子（FNO）的新框架，以提高其学习高频信息和物理性质的能力。SirenFNO利用正弦表示网络学习全网格频谱，无需频率截断，以更少的参数优于标准FNO。GENERIC-FNO将非平衡热力学的能量守恒和熵产生直接嵌入函数空间，展示了精确的结构保证和在各种PDE上的竞争性性能。此外，傅里叶多组分和多层神经网络（FMMNNs）将正弦型激活与多组分结构相结合，实现了指数级的表达能力和有利于高频目标的优化格局。 AI
影响 神经算子架构的这些进步可能导致对物理现象更准确、更有效的模拟，影响流体动力学和材料科学等领域。
排序理由 多篇研究论文介绍了用于学习复杂物理系统的新型神经网络架构。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
arXiv:2606.11518v1 Announce Type: cross Abstract: Fourier neural operators (FNOs) are effective and efficient surrogates for approximating solutions of PDEs and generalize across discretizations. However, owing to the reliance on frequency truncation to maintain learning efficien…
arXiv:2606.08343v1 Announce Type: new Abstract: We introduce GENERIC-FNO, the first neural operator to embed the full GENERIC (metriplectic) structure of nonequilibrium thermodynamics -- reversible, energy-conserving dynamics and irreversible, entropy-producing dynamics coupled t…
We introduce GENERIC-FNO, the first neural operator to embed the full GENERIC (metriplectic) structure of nonequilibrium thermodynamics -- reversible, energy-conserving dynamics and irreversible, entropy-producing dynamics coupled through the degeneracy conditions -- directly in …
arXiv:2502.18959v3 Announce Type: replace-cross Abstract: The architecture of a neural network and the choice of its activation function are both fundamental to its performance. Equally important is ensuring that these two elements are well matched, as their alignment is key to e…