arXiv subjects

Haowen Zhong

Publications and source records attributed to Haowen Zhong.

12 recordsLinked to original sources

arXiv2026

All You Need is not $Ω_\mathrm{gw}$: Beyond the Mean of the Cross-Correlation Estimator when Searching for an Astrophysical Gravitational-Wave Background

Searches for the stochastic gravitational-wave background (SGWB) using ground-based detectors rely heavily on the cross-correlation estimator $\bar C_f$, whose statistical properties determine the theoretical foundation for inference frameworks built on it. Past analyses usually assume Gaussianity of $\bar C_f$, but this assumption has not been systematically tested for a background of astrophysical origin like the one produced by compact binary coalescences. In this work, we decompose $\bar C_f$ into three physically motivated components: (1) the mean intensity, (2) the geometrical shot noise, and (3) the polarization leakage, and we compute 90\% credible intervals for its cumulants up to fourth order for the binary black hole, binary neutron star, and neutron star-black hole populations by propagating local merger rate uncertainties. We find that for the upcoming LIGO O5 sensitivity, $\bar C_f$ remains effectively Gaussian for all three populations, validating current Gaussian likelihood frameworks. For Cosmic Explorer (CE), however, the $\bar C_f$ produced by binary black holes becomes non-Gaussian, with skewness and excess kurtosis of $0.31^{+0.04}_{-0.03}$ and $1.3^{+0.4}_{-0.3}$, respectively, in the most sensitive band over a one-year observation period. For binary neutron star and neutron star - black hole mergers, $\bar C_f$ remains largely Gaussian even at CE. These results indicate that the Edgeworth expansion provides an adequate leading-order description of the non-Gaussianity expected in the next-generation era, while also motivating the development of more sophisticated statistical methods for a more accurate treatment.

gr-qc↗

arXiv2026

A thorough investigation of cross-correlation estimators for stochastic gravitational-wave background searches in ground-based detector data

Detecting a stochastic gravitational-wave background represents a crucial yet challenging objective within the field of gravitational-wave astronomy. Ground-based detectors currently rely almost exclusively on cross-correlation methods to detect stochastic gravitational-wave background signals. Traditionally, these methods define and optimize a broadband estimator initially constructed in the time domain. However, a growing number of analyses require precise narrowband estimators to accurately characterize the energy density of the underlying signal in specific frequency bins. Transitioning from time-domain broadband estimators to frequency-domain narrowband estimators introduces significant complexities that have not yet been fully explored in the existing literature. In this study, we systematically revisit and rigorously reformulate the cross-correlation method in the frequency domain, explicitly addressing and resolving issues related to non-zero covariances induced by windowing and overlapping of data in the time domain. We provide new expressions for the narrowband estimators and their covariances, which differ from those used in past searches. Fortunately, we show that the expressions that have been widely used in the field nonetheless lead to correct posterior distributions for parameter estimation and correct log-Bayes factors for model selection. By establishing a robust theoretical framework, our work facilitates more accurate and physically insightful interpretations of stochastic gravitational-wave background observations, laying an essential foundation for current and future research in this field.

gr-qc↗

arXiv2026

Importance of Shot Noise in the Search for an Isotropic Stochastic Gravitational-Wave Background with Next Generation Detectors

We investigate the impact of shot noise on the stochastic gravitational wave background generated by binary neutron star mergers, and confirm that the overall background can be significantly influenced by relatively few neighboring, loud events. To mitigate the shot noise, we propose a procedure to remove nearby events by notching them out in the time-frequency domain. Additionally, we quantify the cosmic/sample variance of the resulting background after notching, and we study the deviation between the cross-correlation measurement and the theoretical prediction of the background. Taking both effects into account, we find that the resulting sensitivity loss in the search for an isotropic background formed by binary neutron star mergers is minimal, and is limited to $\lesssim 4\%$ below 40 Hz, and to $\lesssim 1\%$ above 40 Hz.

gr-qc↗

arXiv2025

Two-Step Procedure to Detect Cosmological Gravitational Wave Backgrounds with Next-Generation Terrestrial Gravitational-Wave Detectors

Cosmological gravitational-wave backgrounds are an exciting science target for next-generation ground-based detectors, as they encode invaluable information about the primordial Universe. However, any such background is expected to be obscured by the astrophysical foreground from compact-binary coalescences. We propose a novel framework to detect a cosmological gravitational-wave background in the presence of binary black holes and binary neutron star signals with next-generation ground-based detectors, including Cosmic Explorer and the Einstein Telescope. Our procedure involves first removing all the individually resolved binary black hole signals by notching them out in the time-frequency domain. Then, we perform joint Bayesian inference on the individually resolved binary neutron star signals, the unresolved binary neutron star foreground, and the cosmological background. For a flat cosmological background, we find that we can claim detection at $5\,σ$ level when $Ω_\mathrm{ref}\geqslant 2.7\times 10^{-12}/\sqrt{T_\mathrm{obs}/\mathrm{yr}}$, where $T_\mathrm{obs}$ is the observation time (in years), which is within a factor of $\lesssim2$ from the sensitivity reached in absence of these astrophysical foregrounds.

gr-qc↗

arXiv2025

CoopetitiveV: Leveraging LLM-powered Coopetitive Multi-Agent Prompting for High-quality Verilog Generation

Recent advances in agentic LLMs have demonstrated great capabilities in Verilog code generation. However, existing approaches either use LLM-assisted single-agent prompting or cooperation-only multi-agent learning, which will lead to: (i) Degeneration issue for single-agent learning: characterized by diminished error detection and correction capabilities; (ii) Error propagation in cooperation-only multi-agent learning: erroneous information from the former agent will be propagated to the latter through prompts, which can make the latter agents generate buggy code. In this paper, we propose an LLM-based coopetitive multi-agent prompting framework, in which the agents cannot collaborate with each other to form the generation pipeline, but also create a healthy competitive mechanism to improve the generating quality. Our experimental results show that the coopetitive multi-agent framework can effectively mitigate the degeneration risk and reduce the error propagation while improving code error correction capabilities, resulting in higher quality Verilog code generation. The effectiveness of our approach is validated through extensive experiments. On VerilogEval Machine and Human dataset, CoopetitiveV+GPT-4 achieves 99.2% and 99.1% pass@10 scores, respectively. While on RTLLM, CoopetitiveV+GPT-4 obtains 100% syntax and 99.9% functionality pass@5 scores.

cs.LG↗

arXiv2024

Searching for cosmological stochastic backgrounds by notching out resolvable compact binary foregrounds with next-generation gravitational-wave detectors

Stochastic gravitational-wave backgrounds can be of either cosmological or astrophysical origin. The detection of an astrophysical stochastic gravitational-wave background with ground-based interferometers is expected in the near future. Perhaps even more excitingly, the detection of stochastic backgrounds of cosmological origin by future ground-based interferometers could reveal invaluable information about the early Universe. From this perspective, the astrophysical background is a {\it foreground} that can prevent the extraction of this information from the data. In this paper, we revisit a time-frequency domain notching procedure previously proposed to remove the astrophysical foreground in the context of next-generation ground-based detectors, but we consider the more realistic scenario where we remove individually detectable signals by taking into account the uncertainty in the estimation of their parameters. We find that time-frequency domain masks can still efficiently remove the astrophysical foreground and suppress it to about $5\%$ of its original level. Further removal of the foreground formed by unresolvable events (in particular, unresolvable binary neutron stars), which is about $10$ times larger than the residual foreground from realistic notching, would require detector sensitivity improvements. Therefore, the main limitation in the search for a cosmological background is the unresolvable foreground itself, and not the residual of the notching procedure.

gr-qc↗

arXiv2024

Multidimensional hierarchical tests of general relativity with gravitational waves

Tests of general relativity with gravitational waves typically introduce parameters for putative deviations and combine information from multiple events by characterizing the population distribution of these parameters through a hierarchical model. Although many tests include multiple such parameters, hierarchical tests have so far been unable to accommodate this multidimensionality, instead restricting to separate one-dimensional analyses and discarding information about parameter correlations. In this paper, we extend population tests of general relativity to handle an arbitrary number of dimensions. We demonstrate this framework on the two-dimensional inspiral-merger-ringdown consistency test, and derive new constraints from the latest LIGO-Virgo-KAGRA catalog, GWTC-3. We obtain joint constraints for the two parameters introduced by the classic formulation of this test, revealing their correlation structure both at the individual-event and population levels. We additionally propose a new four-dimensional formulation of the inspiral-merger-ringdown test that we show contains further information. As in past work, we find the GW190814 event to be an outlier; the 4D analysis yields further insights on how the low mass and spin of this event biases the population results. Without (with) this event, we find consistency with general relativity at the 60\% (92\%) credible level in the 2D formulation, and 76\% (80\%) for the 4D formulation. This multi-dimensional framework can be immediately applied to other tests of general relativity in any number of dimensions, including the parametrized post-Einsteinian tests and ringdown tests.

gr-qc↗

arXiv2023

pygwb: Python-based library for gravitational-wave background searches

The collection of gravitational waves (GWs) that are either too weak or too numerous to be individually resolved is commonly referred to as the gravitational-wave background (GWB). A confident detection and model-driven characterization of such a signal will provide invaluable information about the evolution of the Universe and the population of GW sources within it. We present a new, user-friendly Python--based package for gravitational-wave data analysis to search for an isotropic GWB in ground--based interferometer data. We employ cross-correlation spectra of GW detector pairs to construct an optimal estimator of the Gaussian and isotropic GWB, and Bayesian parameter estimation to constrain GWB models. The modularity and clarity of the code allow for both a shallow learning curve and flexibility in adjusting the analysis to one's own needs. We describe the individual modules which make up {\tt pygwb}, following the traditional steps of stochastic analyses carried out within the LIGO, Virgo, and KAGRA Collaboration. We then describe the built-in pipeline which combines the different modules and validate it with both mock data and real GW data from the O3 Advanced LIGO and Virgo observing run. We successfully recover all mock data injections and reproduce published results.

gr-qc↗

arXiv2022

Detecting cosmological gravitational waves background after removal of compact binary coalescences in future gravitational wave detectors

The improved sensitivity of third generation gravitational wave detectors opens the possibility of detecting the primordial cosmological stochastic gravitational wave background (SGWB). Detection of the cosmological SGWB is facing a novel challenge: it will likely be masked by the foreground generated by a large number of coalescences of compact binary systems consisting of black holes and/or neutron stars. In this paper, we investigate the possibility of reducing this foreground by removing (notching) the individually resolved compact binary signals in time-frequency space. We establish that such an approach could be used to reach the SGWB sensitivity floor defined by the unresolved part of the compact binaries foreground, which we find to be between $Ω_{\rm GW} \sim (9.1 \times10^{-12} - 8.6\times10^{-11})$ for a frequency independent energy density spectrum and depending on the rate of coalescing binary neutron star systems. Since third-generation gravitational wave detectors will not be able to resolve all compact binaries, the unresolvable component of the compact binaries foreground may limit the SGWB searches with these detectors.

gr-qc↗

arXiv2022

Learning Unbiased Transferability for Domain Adaptation by Uncertainty Modeling

Domain adaptation (DA) aims to transfer knowledge learned from a labeled source domain to an unlabeled or a less labeled but related target domain. Ideally, the source and target distributions should be aligned to each other equally to achieve unbiased knowledge transfer. However, due to the significant imbalance between the amount of annotated data in the source and target domains, usually only the target distribution is aligned to the source domain, leading to adapting unnecessary source specific knowledge to the target domain, i.e., biased domain adaptation. To resolve this problem, in this work, we delve into the transferability estimation problem in domain adaptation and propose a non-intrusive Unbiased Transferability Estimation Plug-in (UTEP) by modeling the uncertainty of a discriminator in adversarial-based DA methods to optimize unbiased transfer. We theoretically analyze the effectiveness of the proposed approach to unbiased transferability learning in DA. Furthermore, to alleviate the impact of imbalanced annotated data, we utilize the estimated uncertainty for pseudo label selection of unlabeled samples in the target domain, which helps achieve better marginal and conditional distribution alignments between domains. Extensive experimental results on a high variety of DA benchmark datasets show that the proposed approach can be readily incorporated into various adversarial-based DA methods, achieving state-of-the-art performance.

cs.CV↗

arXiv2022

Gravitational waves from bubble collisions in FLRW spacetime

Stochastic gravitational wave background (SGWB) is a promising tool to probe the very early universe where the standard model of particle physics and cosmology are connected closely. As a possible component of SGWB, gravitational waves (GW) from bubble collisions during the first order cosmological phase transitions deserve comprehensive analyses. In 2017, Ryusuke Jinno and Masahiro Takimoto proposed an elegant analysis approach to derive the analytical expressions of energy spectra of GW from bubble collisions in Minkowski spacetime avoiding large-scale numerical simulations for the first time[26]. However, they neglect the expansion of the universe and regard the duration of phase transitions as infinity in their derivation which could deviate their estimations from true values. For these two reasons, we give a new expression of GW spectra by adopting their method, switching spacetime background to FLRW spacetime, and considering a finite duration of phase transitions. By denoting $σ$ as the fraction of the speed of phase transitions to the expansion speed of the universe, we find when $σ$ is around $\mathcal{O}(10)$, the maxima of estimated GW energy spectra drop by around 1 order of magnitude than the results given by their previous work. Even when $σ=100$, the maximum of GW energy spectrum is only $65\%$ of their previous estimation. Such a significant decrease may bring about new challenges for the detectability of GW from bubble collisions. Luckily, by comparing new spectra with PLI (\textit{power-law integrated}) sensitivity curves of GW detectors, we find that the detection prospect for GW from bubble collisions is still promising for DECIGO, BBO, LISA, and TianQin in the foreseeable future.

gr-qc↗

arXiv2021

Self-Adaptive Partial Domain Adaptation

Partial Domain adaptation (PDA) aims to solve a more practical cross-domain learning problem that assumes target label space is a subset of source label space. However, the mismatched label space causes significant negative transfer. A traditional solution is using soft weights to increase weights of source shared domain and reduce those of source outlier domain. But it still learns features of outliers and leads to negative immigration. The other mainstream idea is to distinguish source domain into shared and outlier parts by hard binary weights, while it is unavailable to correct the tangled shared and outlier classes. In this paper, we propose an end-to-end Self-Adaptive Partial Domain Adaptation(SAPDA) Network. Class weights evaluation mechanism is introduced to dynamically self-rectify the weights of shared, outlier and confused classes, thus the higher confidence samples have the more sufficient weights. Meanwhile it can eliminate the negative transfer caused by the mismatching of label space greatly. Moreover, our strategy can efficiently measure the transferability of samples in a broader sense, so that our method can achieve competitive results on unsupervised DA task likewise. A large number of experiments on multiple benchmarks have demonstrated the effectiveness of our SAPDA.

cs.CV↗