arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.01368astro-ph.CO
收藏

利用随机置换改进用于宇宙学场的卷积神经网络

Improving Convolutional Neural Networks for Cosmological Fields with Random Permutation

Kunhao Zhong, Marco Gatti, Bhuvnesh Jain

AI总结:

针对低信噪比宇宙学透镜质量图,提出在CNN深层混合池化与随机像素置换作为正则化和数据增强方法,并通过SBI验证其显著提升$S_8$参数约束。

摘要:

卷积神经网络(CNNs)近来已被应用于宇宙学场——弱引力透镜质量图和星系图。然而,宇宙学图在若干方面不同于CNNs已测试过的绝大多数图像:它们是随机的,通常每个像素的信噪比很低,并且在所有尺度上都存在相关性。此外,宇宙学目标是一个回归问题,旨在推断必须无偏的参数后验。我们探索了简单的CNN架构,并提出了一种新的正则化和数据增强方法,以提升其在透镜质量图上的性能。我们发现,通过在深层中混合使用池化和像素混洗,可以获得稳健的改进。随机置换在低信噪比区域对网络进行正则化,并有效增强了现有数据。我们使用基于模拟的推断(SBI)来证明该模型优于文献中的CNN设计。对于模拟的Stage-III巡天,我们发现$S_8$参数的约束改善了30%,其中包括内禀对齐等系统不确定性。我们探索了与下一代巡天相对应的各种统计误差,并发现了相当的改进。我们预期我们的方法也将应用于其他宇宙学场,例如星系图或21-cm图。

英文摘要:

Convolutional Neural Networks (CNNs) have recently been applied to cosmological fields -- weak lensing mass maps and galaxy maps. However, cosmological maps differ in several ways from the vast majority of images that CNNs have been tested on: they are stochastic, typically low signal-to-noise per pixel, and with correlations on all scales. Further, the cosmology goal is a regression problem aimed at inferring posteriors on parameters that must be unbiased. We explore simple CNN architectures and present a novel approach of regularization and data augmentation to improve its performance for lensing mass maps. We find robust improvement by using a mixture of pooling and shuffling of the pixels in the deep layers. The random permutation regularizes the network in the low signal-to-noise regime and effectively augments the existing data. We use simulation-based inference (SBI) to show that the model outperforms CNN designs in the literature. We find a 30% improvement in the constraints of the $S_8$ parameter for simulated Stage-III surveys, including systematic uncertainties such as intrinsic alignments. We explore various statistical errors corresponding to next-generation surveys and find comparable improvements. We expect that our approach will have applications to other cosmological fields as well, such as galaxy maps or 21-cm maps.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑