arXiv subjects

Zirui Chen

Publications and source records attributed to Zirui Chen.

53 records · Page 3Linked to original sources

arXiv2025

VBIM-Net: Variational Born Iterative Network for Inverse Scattering Problems

Recently, studies have shown the potential of integrating field-type iterative methods with deep learning (DL) techniques in solving inverse scattering problems (ISPs). In this article, we propose a novel Variational Born Iterative Network, namely, VBIM-Net, to solve the full-wave ISPs with significantly improved structural rationality and inversion quality. The proposed VBIM-Net emulates the alternating updates of the total electric field and the contrast in the variational Born iterative method (VBIM) by multiple layers of subnetworks. We embed the analytical calculation of the contrast variation into each subnetwork, converting the scattered field residual into an approximate contrast variation and then enhancing it by a U-Net, thus avoiding the requirement of matched measurement dimension and grid resolution as in existing approaches. The total field and contrast of each layer's output is supervised in the loss function of VBIM-Net, imposing soft physical constraints on the variables in the subnetworks, which benefits the model's performance. In addition, we design a training scheme with extra noise to enhance the model's stability. Extensive numerical results on synthetic and experimental data both verify the inversion quality, generalization ability, and robustness of the proposed VBIM-Net. This work may provide some new inspiration for the design of efficient field-type DL schemes.

eess.SP↗

arXiv2024

Paleoinspired Vision: From Exploring Colour Vision Evolution to Inspiring Camera Design

The evolution of colour vision is captivating, as it reveals the adaptive strategies of extinct species while simultaneously inspiring innovations in modern imaging technology. In this study, we present a simplified model of visual transduction in the retina, introducing a novel opsin layer. We quantify evolutionary pressures by measuring machine vision recognition accuracy on colour images shaped by specific opsins. Building on this, we develop an evolutionary conservation optimisation algorithm to reconstruct the spectral sensitivity of opsins, enabling mutation-driven adaptations to to more effectively spot fruits or predators. This model condenses millions of years of evolution within seconds on GPU, providing an experimental framework to test long-standing hypotheses in evolutionary biology , such as vision of early mammals, primate trichromacy from gene duplication, retention of colour blindness, blue-shift of fish rod and multiple rod opsins with bioluminescence. Moreover, the model enables speculative explorations of hypothetical species, such as organisms with eyes adapted to the conditions on Mars. Our findings suggest a minimalist yet effective approach to task-specific camera filter design, optimising the spectral response function to meet application-driven demands. The code will be made publicly available upon acceptance.

cs.CV↗

arXiv2024

Universal dimensions of visual representation

Do neural network models of vision learn brain-aligned representations because they share architectural constraints and task objectives with biological vision or because they learn universal features of natural image processing? We characterized the universality of hundreds of thousands of representational dimensions from visual neural networks with varied construction. We found that networks with varied architectures and task objectives learn to represent natural images using a shared set of latent dimensions, despite appearing highly distinct at a surface level. Next, by comparing these networks with human brain representations measured with fMRI, we found that the most brain-aligned representations in neural networks are those that are universal and independent of a network's specific characteristics. Remarkably, each network can be reduced to fewer than ten of its most universal dimensions with little impact on its representational similarity to the human brain. These results suggest that the underlying similarities between artificial and biological vision are primarily governed by a core set of universal image representations that are convergently learned by diverse systems.

q-bio.NC↗

arXiv2024

Physical Origins of Outflowing Cold Clouds in Local Star-forming Dwarf Galaxies

We study the physical origins of outflowing cold clouds in a sample of 14 low-redshift dwarf ($M_{\ast} \lesssim 10^{10}$ $M_{\odot}$) galaxies from the COS Legacy Archive Spectroscopic SurveY (CLASSY) using Keck/ESI data. Outflows are traced by broad (FWHM ~ 260 $\rm{km}$ $\rm{s^{-1}}$) and very-broad (VB; FWHM ~ 1200 $\rm{km}$ $\rm{s^{-1}}$) velocity components in strong emission lines like [O III] $λ5007$ and $\rm{H}α$. The maximum velocities ($v_{\rm{max}}$) of broad components correlate positively with SFR, unlike the anti-correlation observed for VB components, and are consistent with superbubble models. In contrast, supernova-driven galactic wind models better reproduce the $v_{\rm{max}}$ of VB components. Direct radiative cooling from a hot wind significantly underestimates the luminosities of both broad and VB components. A multi-phase wind model with turbulent radiative mixing reduces this discrepancy to at least one dex for most VB components. Stellar photoionization likely provides additional energy since broad components lie in the starburst locus of excitation diagnostic diagrams. We propose a novel interpretation of outflow origins in star-forming dwarf galaxies$-$broad components trace expanding superbubble shells, while VB components originate from galactic winds. One-zone photoionization models fail to explain the low-ionization lines ([S II] and [O I]) of broad components near the maximal starburst regime, which two-zone photoionization models with density-bounded channels instead reproduce. These two-zone models indicate anisotropic leakage of Lyman continuum photons through low-density channels formed by expanding superbubbles. Our study highlights extreme outflows ($v_{\rm{max}} \gtrsim 1000$ $\rm{km}$ $\rm{s^{-1}}$) in 9 out of 14 star-forming dwarf galaxies, comparable to AGN-driven winds.

astro-ph.GA↗

arXiv2024

Channel Deduction: A New Learning Framework to Acquire Channel from Outdated Samples and Coarse Estimate

How to reduce the pilot overhead required for channel estimation? How to deal with the channel dynamic changes and error propagation in channel prediction? To jointly address these two critical issues in next-generation transceiver design, in this paper, we propose a novel framework named channel deduction for high-dimensional channel acquisition in multiple-input multiple-output (MIMO)-orthogonal frequency division multiplexing (OFDM) systems. Specifically, it makes use of the outdated channel information of past time slots, performs coarse estimation for the current channel with a relatively small number of pilots, and then fuses these two information to obtain a complete representation of the present channel. The rationale is to align the current channel representation to both the latent channel features within the past samples and the coarse estimate of current channel at the pilots, which, in a sense, behaves as a complementary combination of estimation and prediction and thus reduces the overall overhead. To fully exploit the highly nonlinear correlations in time, space, and frequency domains, we resort to learning-based implementation approaches. By using the highly efficient complex-domain multilayer perceptron (MLP)-mixer for crossing space-frequency domain representation and the recurrence-based or attention-based mechanisms for the past-present interaction, we respectively design two different channel deduction neural networks (CDNets). We provide a general procedure of data collection, training, and deployment to standardize the application of CDNets. Comprehensive experimental evaluations in accuracy, robustness, and efficiency demonstrate the superiority of the proposed approach, which reduces the pilot overhead by up to 88.9% compared to state-of-the-art estimation approaches and enables continuous operating even under unknown user movement and error propagation.

eess.SP↗

arXiv2024

Recent Surge in Public Interest in Transportation: Sentiment Analysis of Baidu Apollo Go Using Weibo Data

Urban mobility and transportation systems have been profoundly transformed by the advancement of autonomous vehicle technologies. Baidu Apollo Go, a pioneer robotaxi service from the Chinese tech giant Baidu, has recently been widely deployed in major cities like Beijing and Wuhan, sparking increased conversation and offering a glimpse into the future of urban mobility. This study investigates public attitudes towards Apollo Go across China using Sentiment Analysis with a hybrid BERT model on 36,096 Weibo posts from January to July 2024. The analysis shows that 89.56\% of posts related to Apollo Go are clustered in July. From January to July, public sentiment was mostly positive, but negative comments began to rise after it became a hot topic on July 21. Spatial analysis indicates a strong correlation between provinces with high discussion intensity and those where Apollo Go operates. Initially, Hubei and Guangdong dominated online posting volume, but by July, Guangdong, Beijing, and international regions had overtaken Hubei. Attitudes varied significantly among provinces, with Xinjiang and Qinghai showing optimism and Tibet and Gansu expressing concerns about the impact on traditional taxi services. Sentiment analysis revealed that positive comments focused on technology applications and personal experiences, while negative comments centered on job displacement and safety concerns. In summary, this study highlights the divergence in public perceptions of autonomous ride-hailing services, providing valuable insights for planners, policymakers, and service providers. The model is published on Hugging Face at https://huggingface.co/wsqstar/bert-finetuned-weibo-luobokuaipao and the repository on GitHub at https://github.com/GIStudio/trb2024.

cs.SI↗

arXiv2024

The Survival and Entrainment of Molecules and Dust in Galactic Winds

Recent years have seen excellent progress in modeling the entrainment of T $\sim$ $10^4$K atomic gas in galactic winds. However, the entrainment of cool, dusty T $\sim$ 10-100K molecular gas, which is also observed outflowing at high velocity, is much less understood. Such gas, which can be $10^5$ times denser than the hot wind, appears extremely difficult to entrain. We run 3D wind-tunnel simulations with photoionization self-shielding and evolve thermal dust sputtering and growth. Unlike almost all such simulations to date, we do not enforce any artificial temperature floor. We find efficient molecular gas formation and entrainment, as well as dust survival and growth through accretion. Key to this success is the formation of large amounts of 10^4K atomic gas via mixing, which acts as a protective "bubble wrap" and reduces the cloud overdensity to $\sim$ 100. This can be understood from the ratio of the mixing to cooling time. Before entrainment, when shear is large, t_mix/t_cool $\leq$ 1, and gas cannot cool below the "cooling bottleneck" at 5000K. Thus, the cloud survival criterion is identical to the well-studied purely atomic case. After entrainment, when shear falls, t_mix/t_cool > 1, and the cloud becomes multi-phase, with comparable molecular and atomic masses. The broad temperature PDF, with abundant gas in the formally unstable 50 K < T < 5000 K range, agrees with previous ISM simulations with driven turbulence and radiative cooling. Our findings have implications for dusty molecular gas in stellar and AGN outflows, cluster filaments, "jellyfish" galaxies and AGB winds.

astro-ph.GA↗

arXiv2024

Big AI Models for 6G Wireless Networks: Opportunities, Challenges, and Research Directions

Recently, big artificial intelligence models (BAIMs) represented by chatGPT have brought an incredible revolution. With the pre-trained BAIMs in certain fields, numerous downstream tasks can be accomplished with only few-shot or even zero-shot learning and exhibit state-of-the-art performances. As widely envisioned, the big AI models are to rapidly penetrate into major intelligent services and applications, and are able to run at low unit cost and high flexibility. In 6G wireless networks, to fully enable intelligent communication, sensing and computing, apart from providing other intelligent wireless services and applications, it is of vital importance to design and deploy certain wireless BAIMs (wBAIMs). However, there still lacks investigation on architecture design and system evaluation for wBAIM. In this paper, we provide a comprehensive discussion as well as some in-depth prospects on the demand, design and deployment aspects of the wBAIM. We opine that wBAIM will be a recipe of the 6G wireless networks to build high-efficient, sustainable, versatile, and extensible wireless intelligence for numerous promising visions. Then, we provide the core characteristics, principles, and pilot studies to guide the design of wBAIMs, and discuss the key aspects of developing wBAIMs through identifying the differences between the existing BAIMs and the emerging wBAIMs. Finally, related research directions and potential solutions are outlined.

cs.NI↗

arXiv2024

Channel Mapping Based on Interleaved Learning with Complex-Domain MLP-Mixer

In multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems, representing the whole channel only based on partial subchannels will significantly reduce the channel acquisition overhead. For such a channel mapping task, inspired by the intrinsic coupling across the space and frequency domains, this letter proposes to use interleaved learning with partial antenna and subcarrier characteristics to represent the whole MIMO-OFDM channel. Specifically, we design a complex-domain multilayer perceptron (MLP)-Mixer (CMixer), which utilizes two kinds of complex-domain MLP modules to learn the space and frequency characteristics respectively and then interleaves them to couple the learned properties. The complex-domain computation facilitates the learning on the complex-valued channel data, while the interleaving tightens the coupling of space and frequency domains. These two designs jointly reduce the learning burden, making the physics-inspired CMixer more effective on channel representation learning than existing data-driven approaches. Simulation shows that the proposed scheme brings 4.6~10dB gains in mapping accuracy compared to existing schemes under different settings. Besides, ablation studies show the necessity of complex-domain computation as well as the extent to which the interleaved learning matches the channel properties.

eess.SP↗

arXiv2024

An Edge-Cloud Collaboration Framework for Generative AI Service Provision with Synergetic Big Cloud Model and Small Edge Models

Generative artificial intelligence (GenAI) offers various services to users through content creation, which is believed to be one of the most important components in future networks. However, training and deploying big artificial intelligence models (BAIMs) introduces substantial computational and communication overhead.This poses a critical challenge to centralized approaches, due to the need of high-performance computing infrastructure and the reliability, secrecy and timeliness issues in long-distance access of cloud services. Therefore, there is an urging need to decentralize the services, partly moving them from the cloud to the edge and establishing native GenAI services to enable private, timely, and personalized experiences. In this paper, we propose a brand-new bottom-up BAIM architecture with synergetic big cloud model and small edge models, and design a distributed training framework and a task-oriented deployment scheme for efficient provision of native GenAI services. The proposed framework can facilitate collaborative intelligence, enhance adaptability, gather edge knowledge and alleviate edge-cloud burden. The effectiveness of the proposed framework is demonstrated through an image generation use case. Finally, we outline fundamental research directions to fully exploit the collaborative potential of edge and cloud for native GenAI and BAIM applications.

cs.NI↗

arXiv2023

Exploiting Library Vulnerability via Migration Based Automating Test Generation

In software development, developers extensively utilize third-party libraries to avoid implementing existing functionalities. When a new third-party library vulnerability is disclosed, project maintainers need to determine whether their projects are affected by the vulnerability, which requires developers to invest substantial effort in assessment. However, existing tools face a series of issues: static analysis tools produce false alarms, dynamic analysis tools require existing tests and test generation tools have low success rates when facing complex vulnerabilities. Vulnerability exploits, as code snippets provided for reproducing vulnerabilities after disclosure, contain a wealth of vulnerability-related information. This study proposes a new method based on vulnerability exploits, called VESTA (Vulnerability Exploit-based Software Testing Auto-Generator), which provides vulnerability exploit tests as the basis for developers to decide whether to update dependencies. VESTA extends the search-based test generation methods by adding a migration step, ensuring the similarity between the generated test and the vulnerability exploit, which increases the likelihood of detecting potential library vulnerabilities in a project. We perform experiments on 30 vulnerabilities disclosed in the past five years, involving 60 vulnerability-project pairs, and compare the experimental results with the baseline method, TRANSFER. The success rate of VESTA is 71.7\% which is a 53.4\% improvement over TRANSFER in the effectiveness of verifying exploitable vulnerabilities.

cs.SE↗

arXiv2023

From Data-driven Learning to Physics-inspired Inferring: A Novel Mobile MIMO Channel Prediction Scheme Based on Neural ODE

In this paper, we propose an innovative learning-based channel prediction scheme so as to achieve higher prediction accuracy and reduce the requirements of huge amounts and strict sequential format of channel data. Inspired by the idea of the neural ordinary differential equation (Neural ODE), we first prove that the channel prediction problem can be modeled as an ODE problem with a known initial value by analyzing the physical process of electromagnetic wave propagation within a mobile environment. Then, we design a novel physics-inspired spatial channel gradient network (SCGnet), which represents the derivative process of channel varying as a special neural network and can obtain the gradients at any relative displacement needed for the ODE solving. With the SCGnet, the static channel at any location served by the base station is accurately inferred through consecutive propagation and integration. Finally, we design an efficient recurrent positioning algorithm based on some prior knowledge of user mobility to obtain the velocity vector and propose an approximate Doppler compensation method to make up the instantaneous angular-delay domain channel. Only discrete historical channel data is needed for the training, whereas only a few fresh channel measurements are needed for the prediction, which ensures the scheme's practicability. Comprehensive evaluations show that the proposed scheme is most efficient in representing, learning, and predicting mobile wireless channels.

eess.SP↗

arXiv2023

Non-singular Cooperative Guiding Vector Field Under a Homotopy Equivalence Transformation

The present article advances the concept of a non-singular cooperative guiding vector field under a homotopy equivalence transformation. Firstly, the derivation of a guiding vector field, based on a non-singular vector field, is elaborated to navigate a transformed path from another frame. The existence of such vector fields is also deliberated herein. Subsequently, a coordination vector field derived from the guiding vector field is presented, incorporating an in-depth analysis concerning the impact of the vector field parameters. Lastly, the practical implementation of this novel vector field is demonstrated by its applications to 2-D and 3-D cooperative moving path following issues, establishing its efficacy.

eess.SY↗

arXiv2023

A Frequency-Domain Path-Following Method for Discrete Data-Based Paths

This paper presents a novel frequency-domain approach for path following problem, specifically designed to handle paths described by discrete data. The proposed algorithm utilizes the fast Fourier Transform (FFT) to process the discrete path data, enabling the construction of a non-singular guiding vector field. This vector field serves as a reference direction for the controlled robot, offering the ability to adapt to different levels of precision. Additionally, the frequency-domain nature of the vector field allows for the reduction of computational complexity and effective noise suppression. The efficacy of the proposed approach is demonstrated through a numerical simulation, and theoretical analysis provides an upper bound for the ultimate mean-square path-following error.

eess.SY↗

arXiv2023

The Anatomy of a Turbulent Radiative Mixing Layer: Insights from an Analytic Model with Turbulent Conduction and Viscosity

Turbulent Radiative Mixing Layers (TRMLs) form at the interface of cold, dense gas and hot, diffuse gas in motion with each other. TRMLs are ubiquitous in and around galaxies on a variety of scales, including galactic winds and the circumgalactic medium. They host the intermediate temperature gases that are efficient in radiative cooling, thus play a crucial role in controlling the cold gas supply, phase structure, and spectral features of galaxies. In this work, we develop an intuitive analytic 1.5 dimensional model for TRMLs that includes a simple parameterization of the effective turbulent conductivity and viscosity and a piece-wise power-law cooling curve. Our analytic model reproduces the mass flux, total cooling, and phase structure of 3D simulations of TRMLs at a fraction of the computational cost. It also reveals essential insights into the physics of TRMLs, particularly the importance of the viscous dissipation of relative kinetic energy in balancing radiative cooling as the shear Mach number approaches unity. This dissipation takes place both in the intermediate temperature phase, which reduces the enthalpy flux from the hot phase, and in the cold phase, which enhances radiative cooling. Additionally, our model provides a fast and easy way of computing the column density and surface brightness of TRMLs, which can be directly linked to observations.

astro-ph.GA↗

arXiv2022

Mobile MIMO Channel Prediction with ODE-RNN: a Physics-Inspired Adaptive Approach

Obtaining accurate channel state information (CSI) is crucial and challenging for multiple-input multiple-output (MIMO) wireless communication systems. Conventional channel estimation method cannot guarantee the accuracy of mobile CSI while requires high signaling overhead. Through exploring the intrinsic correlation among a set of historical CSI instances randomly obtained in a certain communication environment, channel prediction can significantly increase CSI accuracy and save signaling overhead. In this paper, we propose a novel channel prediction method based on ordinary differential equation (ODE)-recurrent neural network (RNN) for accurate and flexible mobile MIMO channel prediction. Differing from existing works using sequential network structures for exploring the numerical correlation between observed data, our proposed method tries to represent the implicit physics process of path responses changing by specially designed continuous learning network with ODE structure. Due to the targeted design of learning network, our proposed method fits the mathematics feature of CSI data better and enjoy higher network interpretability. Experimental results show that the proposed learning approach outperforms existing methods, especially for long time interval of the CSI sequence and large channel measurement error.

eess.SP↗

arXiv2022

The Number of Transits Per Epoch for Transiting Misaligned Circumbinary Planets

The growing catalog of circumbinary planets strengthens the notion that planets form in a diverse range of conditions across the cosmos. Transiting circumbinary planets yield especially important insights and many examples are now known, in broadly coplanar obits with respect to their binary. Studies of circumbinary disks suggest misaligned transiting examples could also plausibly exist, but their existence would exacerbate the already challenging feat of automatic detection. In this work, we synthesize populations of such planets and consider the number of transits per epoch they produce, forming integer sequences. For isotropic distributions, such sequences will appear foreign to conventional expectation, rarely (~1%) producing the signature double-transits we have come to expect for circumbinaries, instead producing sparse sequences dominated by zero-transit epochs (~80%). Despite their strangeness, we demonstrate that these sequences will be non-random and that the two preceding epochs predict the next to high accuracy. Additionally, we show that even when clustering the transits into grouped epochs, they often appear unphysical if erroneously assuming a single star, due to the missing epochs. Crucially, missing epochs mean highly isotropic populations can trick the observer into assigning the wrong period in up to a quarter of cases, adding further confusion. Finally, we show that the transit sequences encode the inclination distribution and demonstrate a simple inference method that successfully matches the injected truth. Our work highlights how the simple act of flagging transits can be used to provide an initial, vetting-level analysis of misaligned transiting circumbinary planets.

astro-ph.EP↗