Ermolaev Andrei V, Finot Christophe, Genty Goëry, Dudley John M
Opt Lett. 2024 Aug 1;49(15):4202-4205. doi: 10.1364/OL.524529.
Identifying the underlying processes that locally dominate physical interactions is the key to understanding nonlinear dynamics. Machine-learning techniques have recently been shown to be highly promising in automating the search for dominant physics, adding important insights that complement analytical methods and empirical intuition. Here we apply a fully unsupervised approach to the search for dominant balance during nonlinear and dispersive propagation in an optical fiber and show that we can algorithmically identify dominant interactions in cases of optical wavebreaking, soliton fission, dispersive wave generation, and Raman soliton emergence. We discuss how dominant balance manifests both in the temporal and spectral domains.
识别局部主导物理相互作用的潜在过程是理解非线性动力学的关键。机器学习技术最近已被证明在自动搜索主导物理方面非常有前景，它能提供重要的见解，对分析方法和经验直觉起到补充作用。在此，我们应用一种完全无监督的方法来搜索光纤中非线性和色散传播过程中的主导平衡，并表明我们能够通过算法识别光波破碎、孤子裂变、色散波产生和拉曼孤子出现等情况下的主导相互作用。我们讨论了主导平衡在时域和频域中是如何体现的。
