Evaluating machine learning techniques for predicting power spectra from reionization simulations
Evaluating machine learning techniques for predicting power spectra from reionization simulations
复制标题
DOI：
10.1093/mnras/sty3168
复制
发表时间：
2018-11
影响因子：
4.8
通讯作者：
W. D. Jennings;C. Watkinson;F. Abdalla;J. McEwen
中科院分区：
文献类型：
--
作者：
W. D. Jennings;C. Watkinson;F. Abdalla;J. McEwen
即将进行的实验，如SKA将提供大量的数据。高红移21cm信号的快速建模对于有效地将这些数据集与理论进行比较至关重要。目前最详细的理论预测来自数值模拟和更快但不太准确的半数值模拟。最近，人们提出了机器学习技术来模拟这些半数值模拟的行为，大大减少了时间和计算成本。我们比较了五种这样的机器学习技术的可行性，以模拟公开可用代码SimFast21的21cm功率谱。我们最好的模拟器是一个具有三个隐藏层的多层感知器，它再现SimFast21功率谱的速度比模拟快10^8美元，所有红移和输入参数的平均均方误差为4%。其他技术（插值、高斯过程回归和支持向量机）的预测时间比多层感知器慢，预测精度也比多层感知器差。我们所有的仿真器都可以在任何红移和尺度下进行预测，这提供了更灵活的预测，但在较低的红移下导致预测精度显着降低。然后，我们提出了一种概念验证技术，用于在两个不同的模拟之间进行映射，利用我们最好的模拟器的快速预测速度。我们演示这种技术来查找SimFast21和另一个公开可用的代码21cmFAST之间的映射。我们观察到输入空间某些区域的模拟之间存在明显的偏移。这些技术可能被用作快速半数值模拟和精确数值辐射传输模拟之间的桥梁。
Upcoming experiments such as the SKA will provide huge quantities of data. Fast modelling of the high-redshift 21cm signal will be crucial for efficiently comparing these data sets with theory. The most detailed theoretical predictions currently come from numerical simulations and from faster but less accurate semi-numerical simulations. Recently, machine learning techniques have been proposed to emulate the behaviour of these semi-numerical simulations with drastically reduced time and computing cost. We compare the viability of five such machine learning techniques for emulating the 21cm power spectrum of the publicly-available code SimFast21. Our best emulator is a multilayer perceptron with three hidden layers, reproducing SimFast21 power spectra $10^8$ times faster than the simulation with 4% mean squared error averaged across all redshifts and input parameters. The other techniques (interpolation, Gaussian processes regression, and support vector machine) have slower prediction times and worse prediction accuracy than the multilayer perceptron. All our emulators can make predictions at any redshift and scale, which gives more flexible predictions but results in significantly worse prediction accuracy at lower redshifts. We then present a proof-of-concept technique for mapping between two different simulations, exploiting our best emulator's fast prediction speed. We demonstrate this technique to find a mapping between SimFast21 and another publicly-available code 21cmFAST. We observe a noticeable offset between the simulations for some regions of the input space. Such techniques could potentially be used as a bridge between fast semi-numerical simulations and accurate numerical radiative transfer simulations.