arXiv subjects

Shulei Ni

Publications and source records attributed to Shulei Ni.

11 recordsLinked to original sources

arXiv2026

Machine Learning and the SKA for Cosmic Dawn and the Epoch of Reionization

When operational, the SKA will generate unprecedented amounts of data and provide exquisite sensitivity for 21 cm tomography of Cosmic Dawn (CD) and the Epoch of Reionization (EoR). With this comes opportunities for new data-driven algorithms that unlock new methods for instrument modelling, data analysis, theoretical simulation, and inference for understanding the high-redshift universe. In this chapter, we provide an overview of some machine learning algorithms that have been proposed for CD and EoR science with the SKA

astro-ph.IM↗

arXiv2026

SwinYNet: A Transformer-based Multi-Task Model for Accurate and Efficient FRB Search

In this study, we present a transformer-based multi-task model for Fast Radio Burst (FRB) detection, signal segmentation, and parameter estimation directly from time-frequency data, without requiring computationally expensive de-dispersion preprocessing. To overcome the scarcity of labeled observational data, we develop an FRB simulator and a rule-based automatic annotation pipeline, enabling training exclusively on simulated data. Evaluations on the FAST-FREX dataset show that our model achieves an F1 score of 97.8%, recall of 95.7%, and precision of 100%, outperforming both conventional tools (e.g., PRESTO, Heimdall) and recent AI-based baselines (e.g., RaSPDAM, DRAFTS) in both accuracy and inference speed. The model supports pixel-level signal segmentation and yields reliable estimates for dispersion measure (DM) and time of arrival (ToA). Large-scale blind searches on CRAFTS data further demonstrate robustness, with an average false positive rate of 0.28% and minimal human verification required. This search has already led to the identification of two pulsar candidates, both confirmed as known pulsars. Processing benchmarks indicate that the model enables real-time searches on a single consumer-grade GPU, making petabyte-scale blind searches feasible. The code is publicly available on GitHub, and the model can be easily integrated with existing tools to automate and streamline radio data analysis beyond FRB or pulsar searches.

astro-ph.GA↗

arXiv2025

Deep learning with hybrid frequency differencing and principal component analysis for 21-cm foreground and beam mitigation

Twenty-one-centimeter intensity mapping is a powerful probe of the large-scale distribution of neutral hydrogen (HI) and cosmological observables such as baryon acoustic oscillations. A major challenge is contamination from bright foregrounds and frequency-dependent beam effects, which can lead to signal loss in traditional methods such as principal component analysis (PCA). We develop a hybrid approach that trains a U-shaped convolutional neural network (UNet) on two input channels derived from frequency differencing (FD) and PCA cleaning, enabling it to exploit their complementary behavior across different scales. This two-channel strategy achieves improved performance, maintaining the cross-correlation power spectrum close to unity on large scales under a cosine beam and improving by 5\%-8\% relative to either FD- or PCA-based UNet alone. We further show that the method can robustly recover the HI signal even when the beam model is imperfect and differs between training and testing, with the large-scale cross-correlation remaining close to unity within the $1\sigma$ level. These results demonstrate that the proposed approach provides a robust framework for HI signal reconstruction under realistic observational conditions.

astro-ph.CO↗

arXiv2024

Application of Physics-Informed Neural Networks in Removing Telescope Beam Effects

This study introduces {\tt{PI-AstroDeconv}}, a physics-informed semi-supervised learning method specifically designed for removing beam effects in astronomical telescope observation systems. The method utilizes an encoder-decoder network architecture and combines the telescope's point spread function or beam as prior information, while integrating fast Fourier transform accelerated convolution techniques into the deep learning network. This enables effective removal of beam effects from astronomical observation images. {\tt{PI-AstroDeconv}} can handle multiple PSFs or beams, tolerate imprecise measurements to some extent, and significantly improve the efficiency and accuracy of image deconvolution. Therefore, this architecture is particularly suitable for astronomical data processing that does not rely on annotated data. To validate the reliability of the architecture, we used the SKA Science Data Challenge 3a datasets and compared it with the $\tt{CLEAN}$ deconvolution method at the 21-cm power spectrum level. The results demonstrate that our algorithm not only restores details and reduces blurriness in celestial images at the pixel level but also more accurately recovers the true neutral hydrogen power spectrum at the power spectrum level.

astro-ph.IM↗

arXiv2024

Automated Identification and Segmentation of Hi Sources in CRAFTS Using Deep Learning Method

Identifying neutral hydrogen (\hi) galaxies from observational data is a significant challenge in \hi\ galaxy surveys. With the advancement of observational technology, especially with the advent of large-scale telescope projects such as FAST and SKA, the significant increase in data volume presents new challenges for the efficiency and accuracy of data processing.To address this challenge, in this study, we present a machine learning-based method for extracting \hi\ sources from the three-dimensional (3D) spectral data obtained from the Commensal Radio Astronomy FAST Survey (CRAFTS). We have carefully assembled a specialized dataset, HISF, rich in \hi\ sources, specifically designed to enhance the detection process. Our model, Unet-LK, utilizes the advanced 3D-Unet segmentation architecture and employs an elongated convolution kernel to effectively capture the intricate structures of \hi\ sources. This strategy ensures a reliable identification and segmentation of \hi\ sources, achieving notable performance metrics with a recall rate of 91.6\% and an accuracy of 95.7\%. These results substantiate the robustness of our dataset and the effectiveness of our proposed network architecture in the precise identification of \hi\ sources. Our code and dataset is publicly available at \url{https://github.com/fishszh/HISF}.

cs.CV↗

arXiv2024

PI-AstroDeconv: A Physics-Informed Unsupervised Learning Method for Astronomical Image Deconvolution

In the imaging process of an astronomical telescope, the deconvolution of its beam or Point Spread Function (PSF) is a crucial task. However, deconvolution presents a classical and challenging inverse computation problem. In scenarios where the beam or PSF is complex or inaccurately measured, such as in interferometric arrays and certain radio telescopes, the resultant blurry images are often challenging to interpret visually or analyze using traditional physical detection methods. We argue that traditional methods frequently lack specific prior knowledge, thereby leading to suboptimal performance. To address this issue and achieve image deconvolution and reconstruction, we propose an unsupervised network architecture that incorporates prior physical information. The network adopts an encoder-decoder structure while leveraging the telescope's PSF as prior knowledge. During network training, we introduced accelerated Fast Fourier Transform (FFT) convolution to enable efficient processing of high-resolution input images and PSFs. We explored various classic regression networks, including autoencoder (AE) and U-Net, and conducted a comprehensive performance evaluation through comparative analysis.

astro-ph.IM↗

arXiv2023

CMB delensing with deep learning

The cosmic microwave background (CMB) stands as a pivotal source for studying weak gravitational lensing. While the lensed CMB aids in constraining cosmological parameters, it simultaneously smooths the original CMB's features. The angular power spectrum of the unlensed CMB showcases sharper acoustic peaks and more pronounced damping tails, enhancing the precision of inferring cosmological parameters that influence these aspects. Although delensing diminishes the $B$-mode power spectrum, it facilitates the pursuit of primordial gravitational waves and enables a lower variance reconstruction of lensing and additional sources of secondary CMB anisotropies. In this work, we explore the potential of deep learning techniques, specifically the U-Net++ algorithm, to play a pivotal role in CMB delensing. We analyze three fields, namely $T$, $Q$, and $U$ sky maps, present the angular power spectra of the CMB delensed $TT$, $EE$, $BB$, and $TE$, and compare them with the unlensed CMB angular power spectra. Our findings reveal that the angular power spectrum of the lensed CMB, processed by U-Net++, closely aligns with that of the unlensed CMB. Thus, U-Net++ based CMB delensing proves to be effective in mitigating the impacts of weak gravitational lensing, paving the way for enhancing the CMB delensing power spectrum in forthcoming CMB experiments. The code utilized for this analysis is available on GitHub.

astro-ph.CO↗

arXiv2022

Eliminating polarization leakage effect for neutral hydrogen intensity mapping with deep learning

The neutral hydrogen (HI) intensity mapping (IM) survey is regarded as a promising approach for cosmic large-scale structure (LSS) studies. A major issue for the HI IM survey is to remove the bright foreground contamination. A key to successfully remove the bright foreground is to well control or eliminate the instrumental effects. In this work, we consider the instrumental effect of polarization leakage and use the U-Net approach, a deep learning-based foreground removal technique, to eliminate the polarization leakage effect. The thermal noise is assumed to be a subdominant factor compared with the polarization leakage for future HI IM surveys and ignored in this analysis. In this method, the principal component analysis (PCA) foreground subtraction is used as a preprocessing step for the U-Net foreground subtraction. Our results show that the additional U-Net processing could either remove the foreground residual after the conservative PCA subtraction or compensate for the signal loss caused by the aggressive PCA preprocessing. Finally, we test the robustness of the U-Net foreground subtraction technique and show that it is still reliable in the case of existing constraint error on HI fluctuation amplitude.

astro-ph.IM↗

arXiv2022

Eliminating Primary Beam Effect in Foreground Subtraction of Neutral Hydrogen Intensity Mapping Survey with Deep Learning

In the neutral hydrogen (HI) intensity mapping (IM) survey, the foreground contamination on the cosmological signals is extremely severe, and the systematic effects caused by radio telescopes themselves further aggravate the difficulties in subtracting foreground. In this work, we investigate whether the deep learning method, concretely the 3D U-Net algorithm here, can play a crucial role in foreground subtraction when considering the systematic effect caused by the telescope's primary beam. We consider two beam models, i.e., the Gaussian beam model as a simple case and the Cosine beam model as a sophisticated case. The traditional principal component analysis (PCA) method is employed as a comparison and, more importantly, as the preprocessing step for the U-Net method to reduce the sky map dynamic range. We find that in the case of the Gaussian beam, the PCA method can effectively clean the foreground. However, the PCA method cannot handle the systematic effect induced by the Cosine beam, and the additional U-Net method can improve the result significantly. In order to show how well the PCA and U-Net methods can recover the HI signals, we also derive the HI angular power spectra, as well as the HI 2D power spectra, after performing the foreground subtractions. It is found that, in the case of Gaussian beam, the concordance with the original HI map using U-Net is better than that using PCA by $27.4\%$, and in the case of Cosine beam, the concordance using U-Net is better than that using PCA by $144.8\%$. Therefore, the U-Net based foreground subtraction can efficiently eliminate the telescope primary beam effect and shed new light on recovering the HI power spectrum for future HI IM experiments.

astro-ph.IM↗

arXiv2021

Confronting Inflation Models with the Coming Observations on Primordial Gravitational Waves

The recent observations from CMB have imposed a very stringent upper-limit on the tensor/scalar ratio $r$ of inflation models, $r < 0.064$, which indicates that the primordial gravitational waves (PGW), even though possible to be detected, should have a power spectrum of a tiny amplitude. However, current experiments on PGW is ambitious to detect such a signal by improving the accuracy to an even higher level. Whatever their results are, it will give us much information about the early Universe, not only from the astrophysical side but also from the theoretical side, such as model building for the early Universe. In this paper, we are interested in analyzing what kind of inflation models can be favored by future observations, starting with a kind of general action offered by the effective field theory (EFT) approach. We show a general form of $r$ that can be reduced to various models, and more importantly, we show how the accuracy of future observations can put constraints on model parameters by plotting the contours in their parameter spaces.

astro-ph.CO↗

arXiv2018

Probing signatures of bounce inflation with current observations

The aim of this paper is to probe the features of the bouncing cosmology with the current observational data. Basing on bounce inflation model, with high derivative term, we propose a general parametrization of primordial power spectrum which includes the typical bouncing parameters, such as bouncing time-scale, and energy scale. By applying Markov Chain Monto Carlo analysis with current data combination of Planck 2015, BAO and JLA, we report the posterior probability distributions of the parameters. We find that, bouncing models can well explain CMB observations, especially the deficit and oscillation on large scale in TT power spectrum.

astro-ph.CO↗