arXiv:2410.14764cs.LGcs.NA2024-10被引 17

用低精度数据+少量高精度数据,训练高精度模型,节省昂贵数据成本。

Multifidelity Kolmogorov-Arnold Networks

  • 利用低、高精度数据相关性,用少量高精度数据训练高精度模型。
  • 在无大量高精度数据时,仍可实现准确且鲁棒的预测。
  • 可提升物理信息模型精度,无需额外训练数据,适合科学计算场景。

我们提出多保真度柯尔莫哥洛夫-阿诺德网络(MFKAN),利用低保真度模型与少量高保真度数据,精准训练高保真度数据的模型。MFKAN通过挖掘高低保真度数据间的相关性,显著减少对昂贵高保真度数据的需求,实现在缺乏大规模高保真度数据集时的准确、鲁棒预测。此外,我们证明MFKAN可提升物理信息引导的柯尔莫哥洛夫-阿诺德网络(PIKAN)的精度,且无需额外训练数据。

原文摘要 · Abstract (English)

We develop a method for multifidelity Kolmogorov-Arnold networks (KANs), which use a low-fidelity model along with a small amount of high-fidelity data to train a model for the high-fidelity data accurately. Multifidelity KANs (MFKANs) reduce the amount of expensive high-fidelity data needed to accurately train a KAN by exploiting the correlations between the low- and high-fidelity data to give accurate and robust predictions in the absence of a large high-fidelity dataset. In addition, we show that multifidelity KANs can be used to increase the accuracy of physics-informed KANs (PIKANs), without the use of training data.

神经网络多保真度物理信息

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。