论文标题
部分可观测时空混沌系统的无模型预测
Black Hole Olympics: Metric Selection from the Event Horizon Telescope and LIGO-VIRGO Observations
论文作者
论文摘要
储层计算是预测湍流的有力工具，其简单的架构具有处理大型系统的计算效率。然而，其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间，然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用，该系统模拟了湍流的若干特征。我们表明，使用径向基函数作为非线性投影器，即使只有部分观测并且不知道控制方程，也能稳健地捕捉复杂的系统非线性。最后，我们表明，当测量稀疏、不完整且带有噪声，甚至控制方程变得不准确时，我们的网络仍然可以产生相当准确的预测，从而为实际湍流系统的无模型预测铺平了道路。
We propose a new paradigm in the field of black hole physics, i.e., with more and more high precision observations, one must make a metric selection to determine which black hole is observationally preferred. In light of the shadow imaging data from the Event Horizon Telescope and gravitational wave measurements from the LIGO-VIRGO collaboration, we attempt to address this issue. Although not finding any obvious preference for a specific black hole metric based on current data, as a presentation of this new research paradigm, we rank these black hole metrics using the Bayesian information criterion. We find that Kerr and Reissner-Nordström tie the first place in the black hole Olympics Games. Interestingly, we give the $2\,σ$ upper bound of the average electronic charge $Q<2.82\times10^{18}$ C for the Reissner-Nordström spacetime, and the $2\,σ$ constraint on the average spin parameter $a=-0.02^{+4.07}_{-4.04}$ m for the Kerr spacetime, which is consistent with the prediction of zero rotation for a distant observer. Future multi-messenger and multi-wavelength observations will enhance the application of this new paradigm into tests of gravitational theories and black holes.
