Search arXiv⌕ Search

arXiv subjects

Xin Wang

Publications and source records attributed to Xin Wang.

At least 1,135 records · Page 63Linked to original sources
arXiv2023

AI-assisted reconstruction of cosmic velocity field from redshift-space spatial distribution of halos

The peculiar velocities of dark matter halos are crucial to study many issues in cosmology and galaxy evolution. In this study, by using the state-of-the-art deep learning technique, a UNet-based neural network, we propose to reconstruct the peculiar velocity field from the redshift-space distribution of dark matter halos. Through a point-to-point comparison and examination of various statistical properties, we demonstrate that, the reconstructed velocity field is in good agreement with the ground truth. The power spectra of various velocity field components, including velocity magnitude, divergence and vorticity, can be successfully recovered when $k\lesssim 1.1$ $h/\rm Mpc$ (the Nyquist frequency of the simulations) at about 80% accuracy. This approach is very promising and presents an alternative method to correct the redshift-space distortions using the measured 3D spatial information of halos. Additionally, for the reconstruction of the momentum field of halos, UNet achieves similar good results. Hence the applications in various aspects of cosmology are very broad, such as correcting redshift errors and improving measurements in the structure of the cosmic web, the kinetic Sunyaev-Zel'dovich effect, BAO reconstruction, etc.

astro-ph.CO↗
arXiv2023

Early results from GLASS-JWST. XX: Unveiling a population of "red-excess'' galaxies in Abell2744 and in the coeval field

We combine JWST/NIRCam imaging and MUSE data to characterize the properties of galaxies in different environmental conditions in the cluster Abell2744 ($z=0.3064$) and in its immediate surroundings. We investigate how galaxy colors, morphology and star forming fractions depend on wavelength and on different parameterizations of environment. Our most striking result is the discovery of a ``red-excess'' population in F200W$-$F444W colors both in the cluster regions and the field. These galaxies have normal F115W$-$F150W colors, but are up to 0.8 mag redder than red sequence galaxies in F200W$-$F444W. They also have rather blue rest frame B$-$V colors. {Galaxies in the field and at the cluster virial radius are overall characterized by redder colors, but galaxies with the largest color deviations are found in the field and in the cluster core. Several results} suggest that mechanisms taking place in these regions might be more effective in producing these colors. Looking at their morphology, many cluster galaxies show signatures consistent with ram pressure stripping, while field galaxies have features resembling interactions and mergers. Our hypothesis is that these galaxies are characterized by dust enshrouded star formation: a JWST/NIRSpec spectrum for one of the galaxies is dominated by a strong PAH at 3.3$μm$, suggestive of dust obscured star formation. Larger spectroscopic samples are needed to understand if the color excess is due exclusively to dust-obscured star formation, and the role of environment in triggering it.

astro-ph.GA↗
arXiv2023

Adversarially Robust Neural Architecture Search for Graph Neural Networks

Graph Neural Networks (GNNs) obtain tremendous success in modeling relational data. Still, they are prone to adversarial attacks, which are massive threats to applying GNNs to risk-sensitive domains. Existing defensive methods neither guarantee performance facing new data/tasks or adversarial attacks nor provide insights to understand GNN robustness from an architectural perspective. Neural Architecture Search (NAS) has the potential to solve this problem by automating GNN architecture designs. Nevertheless, current graph NAS approaches lack robust design and are vulnerable to adversarial attacks. To tackle these challenges, we propose a novel Robust Neural Architecture search framework for GNNs (G-RNA). Specifically, we design a robust search space for the message-passing mechanism by adding graph structure mask operations into the search space, which comprises various defensive operation candidates and allows us to search for defensive GNNs. Furthermore, we define a robustness metric to guide the search procedure, which helps to filter robust architectures. In this way, G-RNA helps understand GNN robustness from an architectural perspective and effectively searches for optimal adversarial robust GNNs. Extensive experimental results on benchmark datasets show that G-RNA significantly outperforms manually designed robust GNNs and vanilla graph NAS baselines by 12.1% to 23.4% under adversarial attacks.

cs.LG↗
arXiv2023

Information recoverability of noisy quantum states

Extracting classical information from quantum systems is an essential step of many quantum algorithms. However, this information could be corrupted as the systems are prone to quantum noises, and its distortion under quantum dynamics has not been adequately investigated. In this work, we introduce a systematic framework to study how well we can retrieve information from noisy quantum states. Given a noisy quantum channel, we fully characterize the range of recoverable classical information. This condition allows a natural measure quantifying the information recoverability of a channel. Moreover, we resolve the minimum information retrieving cost, which, along with the corresponding optimal protocol, is efficiently computable by semidefinite programming. As applications, we establish the limits on the information retrieving cost for practical quantum noises and employ the corresponding protocols to mitigate errors in ground state energy estimation. Our work gives the first full characterization of information recoverability of noisy quantum states from the recoverable range to the recovering cost, revealing the ultimate limit of probabilistic error cancellation.

quant-ph↗
arXiv2023

Early Results from GLASS-JWST XXII: Rest frame UV-optical spectral properties of Lyman-alpha emitting galaxies at 3 $<$ z $<$ 6

Ly$α$ emission is possibly the best indirect diagnostic of Lyman continuum (LyC) escape since the conditions that favor the escape of Ly$α$ photons are often the same that allows for the escape of LyC photons. In this work, we present the rest UV-optical spectral characteristics of 11 Ly$α$ emitting galaxies at 3 $<$ z $<$ 6 - the optimal redshift range chosen to avoid the extreme IGM attenuation while simultaneously studying galaxies close enough to the epoch of reionization. From a combined analysis of JWST/NIRSpec and MUSE data, we present the Ly$α$ escape fraction and study their correlations with other physical properties of galaxies that might facilitate Ly$α$ escape. We find that our galaxies have low masses (80\% of the sample with $\rm log_{10} \ M_{\star} < 9.5\ M_{\odot}$), compact sizes (median $\rm R_e \sim 0.7 \ kpc $), low dust content, moderate [OIII]/[OII] flux ratios (mean $\sim$ 6.8 $\pm$ 1.2), and moderate Ly$α$ escape fraction (mean $\rm f_{esc}^{Lyα} \ \sim$ 0.11). Our sample show characteristics that are broadly consistent with the low redshift galaxies with Ly$α$ emission, which are termed as "analogs" of high redshift population. We predict the Lyman continuum escape fraction in our sample to be low (0.03-0.07), although larger samples in the post-reionization epoch are needed to confirm these trends.

astro-ph.GA↗
arXiv2023

CEERS: Spatially Resolved UV and mid-IR Star Formation in Galaxies at 0.2 < z < 2.5: The Picture from the Hubble and James Webb Space Telescopes

We present the mid-IR (MIR) morphologies for 64 star-forming galaxies at $0.2 10^{9}~M_\odot}$ using JWST MIRI observations from the Cosmic Evolution Early Release Science survey (CEERS). The MIRI bands span the MIR (7.7--21~$μ$m), enabling us to measure the effective radii ($R_{\rm{eff}}$) and Sérsic indexes of these SFGs at rest-frame 6.2 and 7.7 $μ$m, which contains strong emission from Polycyclic aromatic hydrocarbon (PAH) features, a well-established tracer of star formation in galaxies. We define a ``PAH-band'' as the MIRI bandpass that contains these features at the redshift of the galaxy. We then compare the galaxy morphologies in the PAH-bands to those in rest-frame Near-UV (NUV) using HST ACS/F435W or ACS/F606W and optical/near-IR using HST WFC3/F160W imaging from UVCANDELS and CANDELS, where the NUV-band and F160W trace the profile of (unobscured) massive stars and the stellar continuum, respectively. The $R_{\rm{eff}}$ of galaxies in the PAH-band are slightly smaller ($\sim$10\%) than those in F160W for galaxies with $\rm{M_*\gtrsim10^{9.5}~M_\odot}$ at $z\leq1.2$, but the PAH-band and F160W have a similar fractions of light within 1 kpc. In contrast, the $R_{\rm{eff}}$ of galaxies in the NUV-band are larger, with lower fractions of light within 1 kpc compared to F160W for galaxies at $z\leq1.2$. Using the MIRI data to estimate the $\rm{SFR_{\rm{IR}}}$ surface density, we find the correlation between the $\rm{SFR_{\rm{IR}}}$ surface density and stellar mass has a steeper slope than that of the $\rm{SFR_{\rm{UV}}}$ surface density and stellar mass, suggesting more massive galaxies having increasing amounts of obscured fraction of star formation in their inner regions. This paper demonstrates how the high-angular resolution data from JWST/MIRI can reveal new information about the morphology of obscured-star formation.

astro-ph.GA↗
arXiv2023

Blockchain Mining with Multiple Selfish Miners

This paper studies a fundamental problem regarding the security of blockchain PoW consensus on how the existence of multiple misbehaving miners influences the profitability of selfish mining. Each selfish miner (or attacker interchangeably) maintains a private chain and makes it public opportunistically for acquiring more rewards incommensurate to his Hash power. We first establish a general Markov chain model to characterize the state transition of public and private chains for Basic Selfish Mining (BSM), and derive the stationary profitable threshold of Hash power in closed-form. It reduces from 25% for a single attacker to below 21.48% for two symmetric attackers theoretically, and further reduces to around 10% with eight symmetric attackers experimentally. We next explore the profitable threshold when one of the attackers performs strategic mining based on Partially Observable Markov Decision Process (POMDP) that only half of the attributes pertinent to a mining state are observable to him. An online algorithm is presented to compute the nearly optimal policy efficiently despite the large state space and high dimensional belief space. The strategic attacker mines selfishly and more agilely than BSM attacker when his Hash power is relatively high, and mines honestly otherwise, thus leading to a much lower profitable threshold. Last, we formulate a simple model of absolute mining revenue that yields an interesting observation: selfish mining is never profitable at the first difficulty adjustment period, but replying on the reimbursement of stationary selfish mining gains in the future periods. The delay till being profitable of an attacker increases with the decrease of his Hash power, making blockchain miners more cautious on performing selfish mining.

cs.CR↗
arXiv2023

CIMI4D: A Large Multimodal Climbing Motion Dataset under Human-scene Interactions

Motion capture is a long-standing research problem. Although it has been studied for decades, the majority of research focus on ground-based movements such as walking, sitting, dancing, etc. Off-grounded actions such as climbing are largely overlooked. As an important type of action in sports and firefighting field, the climbing movements is challenging to capture because of its complex back poses, intricate human-scene interactions, and difficult global localization. The research community does not have an in-depth understanding of the climbing action due to the lack of specific datasets. To address this limitation, we collect CIMI4D, a large rock \textbf{C}l\textbf{I}mbing \textbf{M}ot\textbf{I}on dataset from 12 persons climbing 13 different climbing walls. The dataset consists of around 180,000 frames of pose inertial measurements, LiDAR point clouds, RGB videos, high-precision static point cloud scenes, and reconstructed scene meshes. Moreover, we frame-wise annotate touch rock holds to facilitate a detailed exploration of human-scene interaction. The core of this dataset is a blending optimization process, which corrects for the pose as it drifts and is affected by the magnetic conditions. To evaluate the merit of CIMI4D, we perform four tasks which include human pose estimations (with/without scene constraints), pose prediction, and pose generation. The experimental results demonstrate that CIMI4D presents great challenges to existing methods and enables extensive research opportunities. We share the dataset with the research community in http://www.lidarhumanmotion.net/cimi4d/.

cs.CV↗
arXiv2023

Domain Adaptive Semantic Segmentation by Optimal Transport

Scene segmentation is widely used in the field of autonomous driving for environment perception, and semantic scene segmentation (3S) has received a great deal of attention due to the richness of the semantic information it contains. It aims to assign labels to pixels in an image, thus enabling automatic image labeling. Current approaches are mainly based on convolutional neural networks (CNN), but they rely on a large number of labels. Therefore, how to use a small size of labeled data to achieve semantic segmentation becomes more and more important. In this paper, we propose a domain adaptation (DA) framework based on optimal transport (OT) and attention mechanism to address this issue. Concretely, first we generate the output space via CNN due to its superiority of feature representation. Second, we utilize OT to achieve a more robust alignment of source and target domains in output space, where the OT plan defines a well attention mechanism to improve the adaptation of the model. In particular, with OT, the number of network parameters has been reduced and the network has been better interpretable. Third, to better describe the multi-scale property of features, we construct a multi-scale segmentation network to perform domain adaptation. Finally, in order to verify the performance of our proposed method, we conduct experimental comparison with three benchmark and four SOTA methods on three scene datasets, and the mean intersection-over-union (mIOU) has been significant improved, and visualization results under multiple domain adaptation scenarios also show that our proposed method has better performance than compared semantic segmentation methods.

cs.CV↗
arXiv2023

Hiding speaker's sex in speech using zero-evidence speaker representation in an analysis/synthesis pipeline

The use of modern vocoders in an analysis/synthesis pipeline allows us to investigate high-quality voice conversion that can be used for privacy purposes. Here, we propose to transform the speaker embedding and the pitch in order to hide the sex of the speaker. ECAPA-TDNN-based speaker representation fed into a HiFiGAN vocoder is protected using a neural-discriminant analysis approach, which is consistent with the zero-evidence concept of privacy. This approach significantly reduces the information in speech related to the speaker's sex while preserving speech content and some consistency in the resulting protected voices.

eess.AS↗
arXiv2023

Top-Down Visual Attention from Analysis by Synthesis

Current attention algorithms (e.g., self-attention) are stimulus-driven and highlight all the salient objects in an image. However, intelligent agents like humans often guide their attention based on the high-level task at hand, focusing only on task-related objects. This ability of task-guided top-down attention provides task-adaptive representation and helps the model generalize to various tasks. In this paper, we consider top-down attention from a classic Analysis-by-Synthesis (AbS) perspective of vision. Prior work indicates a functional equivalence between visual attention and sparse reconstruction; we show that an AbS visual system that optimizes a similar sparse reconstruction objective modulated by a goal-directed top-down signal naturally simulates top-down attention. We further propose Analysis-by-Synthesis Vision Transformer (AbSViT), which is a top-down modulated ViT model that variationally approximates AbS, and achieves controllable top-down attention. For real-world applications, AbSViT consistently improves over baselines on Vision-Language tasks such as VQA and zero-shot retrieval where language guides the top-down attention. AbSViT can also serve as a general backbone, improving performance on classification, semantic segmentation, and model robustness.

cs.CV↗
arXiv2023

Damage detection of high-speed railway box girder using train-induced dynamic responses

This paper proposes a damage detection method based on the train-induced responses of high-speed railway box girder. Under the coupling effects of bending and torsion, the traditional damage detection method based on the Euler beam theory cannot be applied. In this research, the box girder section is divided into different components based on the plate element analysis method. The strain responses were preprocessed based on Principal Component Analysis (PCA) method to remove the influence of train operation variation. The residual error of autoregressive (AR) model was used as a potential index of damage features. The optimal order of the model was determined based on Bayesian information criterion (BIC) criterion. Finally, the confidence boundary (CB) of damage features (DF) constituting outliers can be estimated by Gaussian inverse cumulative distribution function (ICDF). The numerical simulation results show that the proposed method in this paper can effectively identify, locate and quantify the damage, which verifies the accuracy of the proposed method. The proposed method effectively identifies the early damage of all components on the key section by using four strain sensors, and it is helpful for developing effective maintenance strategies for high-speed railway box girder.

eess.SP↗
arXiv2023

Doubly Right Object Recognition: A Why Prompt for Visual Rationales

Many visual recognition models are evaluated only on their classification accuracy, a metric for which they obtain strong performance. In this paper, we investigate whether computer vision models can also provide correct rationales for their predictions. We propose a ``doubly right'' object recognition benchmark, where the metric requires the model to simultaneously produce both the right labels as well as the right rationales. We find that state-of-the-art visual models, such as CLIP, often provide incorrect rationales for their categorical predictions. However, by transferring the rationales from language models into visual representations through a tailored dataset, we show that we can learn a ``why prompt,'' which adapts large visual representations to produce correct rationales. Visualizations and empirical experiments show that our prompts significantly improve performance on doubly right object recognition, in addition to zero-shot transfer to unseen tasks and datasets.

cs.CV↗
arXiv2023

Hierarchical Knowledge Guided Learning for Real-world Retinal Diseases Recognition

In the real world, medical datasets often exhibit a long-tailed data distribution (i.e., a few classes occupy the majority of the data, while most classes have only a limited number of samples), which results in a challenging long-tailed learning scenario. Some recently published datasets in ophthalmology AI consist of more than 40 kinds of retinal diseases with complex abnormalities and variable morbidity. Nevertheless, more than 30 conditions are rarely seen in global patient cohorts. From a modeling perspective, most deep learning models trained on these datasets may lack the ability to generalize to rare diseases where only a few available samples are presented for training. In addition, there may be more than one disease for the presence of the retina, resulting in a challenging label co-occurrence scenario, also known as \textit{multi-label}, which can cause problems when some re-sampling strategies are applied during training. To address the above two major challenges, this paper presents a novel method that enables the deep neural network to learn from a long-tailed fundus database for various retinal disease recognition. Firstly, we exploit the prior knowledge in ophthalmology to improve the feature representation using a hierarchy-aware pre-training. Secondly, we adopt an instance-wise class-balanced sampling strategy to address the label co-occurrence issue under the long-tailed medical dataset scenario. Thirdly, we introduce a novel hybrid knowledge distillation to train a less biased representation and classifier. We conducted extensive experiments on four databases, including two public datasets and two in-house databases with more than one million fundus images. The experimental results demonstrate the superiority of our proposed methods with recognition accuracy outperforming the state-of-the-art competitors, especially for these rare diseases.

cs.CV↗
arXiv2023

Can Knowledge of End-to-End Text-to-Speech Models Improve Neural MIDI-to-Audio Synthesis Systems?

With the similarity between music and speech synthesis from symbolic input and the rapid development of text-to-speech (TTS) techniques, it is worthwhile to explore ways to improve the MIDI-to-audio performance by borrowing from TTS techniques. In this study, we analyze the shortcomings of a TTS-based MIDI-to-audio system and improve it in terms of feature computation, model selection, and training strategy, aiming to synthesize highly natural-sounding audio. Moreover, we conducted an extensive model evaluation through listening tests, pitch measurement, and spectrogram analysis. This work demonstrates not only synthesis of highly natural music but offers a thorough analytical approach and useful outcomes for the community. Our code, pre-trained models, supplementary materials, and audio samples are open sourced at https://github.com/nii-yamagishilab/midi-to-audio.

cs.SD↗
arXiv2023

The GLASS-JWST Early Release Science Program. III. Strong lensing model of Abell 2744 and its infalling regions

We present a new high-precision, JWST-based, strong lensing model for the galaxy cluster Abell 2744 at $z=0.3072$. By combining the deep, high-resolution JWST imaging from the GLASS-JWST and UNCOVER programs and a Director's Discretionary Time program, with newly obtained VLT/MUSE data, we identify 32 multiple images from 11 background sources lensed by two external sub-clusters at distances of ~160" from the main cluster. The new MUSE observations enable the first spectroscopic confirmation of a multiple image system in the external clumps. Moreover, the re-analysis of the spectro-photometric archival and JWST data yields 27 additional multiple images in the main cluster. The new lens model is constrained by 149 multiple images ($\sim66\%$ more than in our previous Bergamini et al. 2023 model) covering an extended redshift range between 1.03 and 9.76. The subhalo mass component of the cluster includes 177 member galaxies down to $m_{\rm F160W}=21$, 163 of which are spectroscopically confirmed. Internal velocity dispersions are measured for 85 members. The new lens model is characterized by a remarkably low scatter between predicted and observed positions of the multiple images (0.43"). This precision is unprecedented given the large multiple image sample, the complexity of the cluster mass distribution, and the large modeled area. The improved accuracy and resolution of the cluster total mass distribution provides a robust magnification map over a $\sim\!45$ arcmin$^2$ area, which is critical for inferring the intrinsic physical properties of the highly magnified, high-$z$ sources. The lens model and the new MUSE redshift catalog are released with this publication.

astro-ph.GA↗
arXiv2023

Spatial Attention-based Implicit Neural Representation for Arbitrary Reduction of MRI Slice Spacing

Magnetic resonance (MR) images collected in 2D clinical protocols typically have large inter-slice spacing, resulting in high in-plane resolution and reduced through-plane resolution. Super-resolution technique can enhance the through-plane resolution of MR images to facilitate downstream visualization and computer-aided diagnosis. However, most existing works train the super-resolution network at a fixed scaling factor, which is not friendly to clinical scenes of varying inter-slice spacing in MR scanning. Inspired by the recent progress in implicit neural representation, we propose a Spatial Attention-based Implicit Neural Representation (SA-INR) network for arbitrary reduction of MR inter-slice spacing. The SA-INR aims to represent an MR image as a continuous implicit function of 3D coordinates. In this way, the SA-INR can reconstruct the MR image with arbitrary inter-slice spacing by continuously sampling the coordinates in 3D space. In particular, a local-aware spatial attention operation is introduced to model nearby voxels and their affinity more accurately in a larger receptive field. Meanwhile, to improve the computational efficiency, a gradient-guided gating mask is proposed for applying the local-aware spatial attention to selected areas only. We evaluate our method on the public HCP-1200 dataset and the clinical knee MR dataset to demonstrate its superiority over other existing methods.

eess.IV↗
arXiv2023

Ground state preparation with shallow variational warm-start

Preparing the ground states of a many-body system is essential for evaluating physical quantities and determining the properties of materials. This work provides a quantum ground state preparation scheme with shallow variational warm-start to tackle the bottlenecks of current algorithms, i.e., demand for prior ground state energy information and lack of demonstration of efficient initial state preparation. Particularly, our methods would not experience the instability for small spectral gap $Δ$ during pre-encoding the phase factors since our methods involve only $\widetilde{O}(1)$ factors while $\widetilde{O}(Δ^{-1})$ is requested by the near-optimal methods. We demonstrate the effectiveness of our methods via extensive numerical simulations on spin-$1/2$ Heisenberg models. We also show that the shallow warm-start procedure can process chemical molecules by conducting numerical simulations on the hydrogen chain model. Moreover, we extend research on the Hubbard model, demonstrating superior performance compared to the prevalent variational quantum algorithms.

quant-ph↗