arXiv subjects

Yi Xu

Publications and source records attributed to Yi Xu.

At least 253 records · Page 14Linked to original sources

arXiv2024

EMPRESS. XIV. Strong High Ionization Lines of Young Galaxies at $z=0-8$: Ionizing Spectra Consistent with the Intermediate Mass Black Holes with $M_{\rm BH}\sim 10^3-10^6\ M_\odot$

We present ionizing spectra estimated at 13.6--100 eV for ten dwarf galaxies with strong high ionization lines of He {\sc {ii}}$λ$4686 and [Ne {\sc{v}}]$λ$3426 ([Ne {\sc{iv}}]$λ$2424) at $z=0$ ($z=8$) that are identified in our Keck/LRIS spectroscopy and the literature (the JWST ERO program). With the flux ratios of these high ionization lines and $>10$ low-ionization lines of hydrogen, helium, oxygen, neon, and sulfur, we determine ionizing spectra consisting of stellar and non-thermal power-law radiation by photoionization modeling with free parameters of nebular properties including metallicity and ionization parameter, cancelling out abundance ratio differences. We find that all of the observed flux ratios are well reproduced by the photoinization models with the power law index $α_{\rm EUV}$ of $α_{\rm EUV}\sim (-1)-0$ and the luminosity $L_{\rm EUV}$ of $L_{\rm EUV}\sim 10^{40}-10^{42}$ erg s$^{-1}$ at $\sim 55-100$ eV for six galaxies, while four galaxies include large systematics in $α_{\rm EUV}$ caused by stellar radiation contamination. We then compare $α_{\rm EUV}$ and $L_{\rm EUV}$ of these six galaxies with those predicted by the black hole (BH) accretion disk models, and find that $α_{\rm EUV}$ and $L_{\rm EUV}$ are similar to those of the intermediate mass black holes (IMBHs) in BH accretion disk models {albeit with possibilities of the other scenarios.} Confirming these results with a known IMBH having a mass $M_{\rm BH}$ of $M_{\rm BH}=10^{5.75} \ M_\odot$, we find that four local galaxies and one $z=7.665$ galaxy have ionizing spectra consistent with those of IMBHs with $M_{\rm BH} \sim 10^3-10^5 \ M_\odot$.

astro-ph.GA↗

arXiv2024

Spectrum AUC Difference (SAUCD): Human-aligned 3D Shape Evaluation

Existing 3D mesh shape evaluation metrics mainly focus on the overall shape but are usually less sensitive to local details. This makes them inconsistent with human evaluation, as human perception cares about both overall and detailed shape. In this paper, we propose an analytic metric named Spectrum Area Under the Curve Difference (SAUCD) that demonstrates better consistency with human evaluation. To compare the difference between two shapes, we first transform the 3D mesh to the spectrum domain using the discrete Laplace-Beltrami operator and Fourier transform. Then, we calculate the Area Under the Curve (AUC) difference between the two spectrums, so that each frequency band that captures either the overall or detailed shape is equitably considered. Taking human sensitivity across frequency bands into account, we further extend our metric by learning suitable weights for each frequency band which better aligns with human perception. To measure the performance of SAUCD, we build a 3D mesh evaluation dataset called Shape Grading, along with manual annotations from more than 800 subjects. By measuring the correlation between our metric and human evaluation, we demonstrate that SAUCD is well aligned with human evaluation, and outperforms previous 3D mesh metrics.

cs.CV↗

arXiv2024

Learning with Semantics: Towards a Semantics-Aware Routing Anomaly Detection System

BGP is the de facto inter-domain routing protocol to ensure global connectivity of the Internet. However, various reasons, such as deliberate attacks or misconfigurations, could cause BGP routing anomalies. Traditional methods for BGP routing anomaly detection require significant manual investigation of routes by network operators. Although machine learning has been applied to automate the process, prior arts typically impose significant training overhead (such as large-scale data labeling and feature crafting), and only produce uninterpretable results. To address these limitations, this paper presents a routing anomaly detection system centering around a novel network representation learning model named BEAM. The core design of BEAM is to accurately learn the unique properties (defined as \emph{routing role}) of each Autonomous System (AS) in the Internet by incorporating BGP semantics. As a result, routing anomaly detection, given BEAM, is reduced to a matter of discovering unexpected routing role churns upon observing new route announcements. We implement a prototype of our routing anomaly detection system and extensively evaluate its performance. The experimental results, based on 18 real-world RouteViews datasets containing over 11 billion route announcement records, demonstrate that our system can detect all previously-confirmed routing anomalies, while only introducing at most five false alarms every 180 million route announcements. We also deploy our system at a large ISP to perform real-world detection for one month. During the course of deployment, our system detects 497 true anomalies in the wild with an average of only 1.65 false alarms per day.

cs.NI↗

arXiv2024

Field induced non-BEC transitions in frustrated magnets

Frustrated spin-systems have traditionally proven challenging to understand, owing to a scarcity of controlled methods for their analyses. By contrast, under strong magnetic fields, certain aspects of spin systems admit simpler and universal description in terms of hardcore bosons. The bosonic formalism is anchored by the phenomenon of Bose-Einstein condensation (BEC), which has helped explain the behaviors of a wide range of magnetic compounds under applied magnetic fields. Here, we focus on the interplay between frustration and externally applied magnetic field to identify instances where the BEC paradigm is no longer applicable. As a representative example, we consider the antiferromagnetic $J_1 - J_2 - J_3$ model on the square lattice in the presence of a uniform external magnetic field, and demonstrate that the frustration-driven suppression of the Néel order leads to a Lifshitz transition for the hardcore bosons. In the vicinity of the Lifshitz point, the physics becomes unmoored from the BEC paradigm, and the behavior of the system, both at and below the saturation field, is controlled by a Lifshitz multicritical point. We obtain the resultant universal scaling behaviors, and provide strong evidence for the existence of a frustration and magnetic-field driven correlated bosonic liquid state along the entire phase boundary separating the Néel phase from other magnetically ordered states.

cond-mat.str-el↗

arXiv2024

OpenIllumination: A Multi-Illumination Dataset for Inverse Rendering Evaluation on Real Objects

We introduce OpenIllumination, a real-world dataset containing over 108K images of 64 objects with diverse materials, captured under 72 camera views and a large number of different illuminations. For each image in the dataset, we provide accurate camera parameters, illumination ground truth, and foreground segmentation masks. Our dataset enables the quantitative evaluation of most inverse rendering and material decomposition methods for real objects. We examine several state-of-the-art inverse rendering methods on our dataset and compare their performances. The dataset and code can be found on the project page: https://oppo-us-research.github.io/OpenIllumination.

cs.CV↗

arXiv2024

Better Representations via Adversarial Training in Pre-Training: A Theoretical Perspective

Pre-training is known to generate universal representations for downstream tasks in large-scale deep learning such as large language models. Existing literature, e.g., \cite{kim2020adversarial}, empirically observe that the downstream tasks can inherit the adversarial robustness of the pre-trained model. We provide theoretical justifications for this robustness inheritance phenomenon. Our theoretical results reveal that feature purification plays an important role in connecting the adversarial robustness of the pre-trained model and the downstream tasks in two-layer neural networks. Specifically, we show that (i) with adversarial training, each hidden node tends to pick only one (or a few) feature; (ii) without adversarial training, the hidden nodes can be vulnerable to attacks. This observation is valid for both supervised pre-training and contrastive learning. With purified nodes, it turns out that clean training is enough to achieve adversarial robustness in downstream tasks.

cs.LG↗

arXiv2024

EMPRESS. XII. Statistics on the Dynamics and Gas Mass Fraction of Extremely-Metal Poor Galaxies

We present demography of the dynamics and gas-mass fraction of 33 extremely metal-poor galaxies (EMPGs) with metallicities of $0.015-0.195~Z_\odot$ and low stellar masses of $10^4-10^8~M_\odot$ in the local universe. We conduct deep optical integral-field spectroscopy (IFS) for the low-mass EMPGs with the medium high resolution ($R=7500$) grism of the 8m-Subaru FOCAS IFU instrument by the EMPRESS 3D survey, and investigate H$α$ emission of the EMPGs. Exploiting the resolution high enough for the low-mass galaxies, we derive gas dynamics with the H$α$ lines by the fitting of 3-dimensional disk models. We obtain an average maximum rotation velocity ($v_\mathrm{rot}$) of $15\pm3~\mathrm{km~s^{-1}}$ and an average intrinsic velocity dispersion ($σ_0$) of $27\pm10~\mathrm{km~s^{-1}}$ for 15 spatially resolved EMPGs out of the 33 EMPGs, and find that all of the 15 EMPGs have $v_\mathrm{rot}/σ_0<1$ suggesting dispersion dominated systems. There is a clear decreasing trend of $v_\mathrm{rot}/σ_0$ with the decreasing stellar mass and metallicity. We derive the gas mass fraction ($f_\mathrm{gas}$) for all of the 33 EMPGs, and find no clear dependence on stellar mass and metallicity. These $v_\mathrm{rot}/σ_0$ and $f_\mathrm{gas}$ trends should be compared with young high-$z$ galaxies observed by the forthcoming JWST IFS programs to understand the physical origins of the EMPGs in the local universe.

astro-ph.GA↗

arXiv2024

2.75D: Boosting learning by representing 3D Medical imaging to 2D features for small data

In medical-data driven learning, 3D convolutional neural networks (CNNs) have started to show superior performance to 2D CNNs in numerous deep learning tasks, proving the added value of 3D spatial information in feature representation. However, the difficulty in collecting more training samples to converge, more computational resources and longer execution time make this approach less applied. Also, applying transfer learning on 3D CNN is challenging due to a lack of publicly available pre-trained 3D models. To tackle these issues, we proposed a novel 2D strategical representation of volumetric data, namely 2.75D. In this work, the spatial information of 3D images is captured in a single 2D view by a spiral-spinning technique. As a result, 2D CNN networks can also be used to learn volumetric information. Besides, we can fully leverage pre-trained 2D CNNs for downstream vision problems. We also explore a multi-view 2.75D strategy, 2.75D 3 channels (2.75Dx3), to boost the advantage of 2.75D. We evaluated the proposed methods on three public datasets with different modalities or organs (Lung CT, Breast MRI, and Prostate MRI), against their 2D, 2.5D, and 3D counterparts in classification tasks. Results show that the proposed methods significantly outperform other counterparts when all methods were trained from scratch on the lung dataset. Such performance gain is more pronounced with transfer learning or in the case of limited training data. Our methods also achieved comparable performance on other datasets. In addition, our methods achieved a substantial reduction in time consumption of training and inference compared with the 2.5D or 3D method.

eess.IV↗

arXiv2024

Census for the Rest-frame Optical and UV Morphologies of Galaxies at $z=4-10$: First Phase of Inside-Out Galaxy Formation

We present the rest-frame optical and UV surface brightness (SB) profiles for $149$ galaxies with $M_{\rm opt}< -19.4$ mag at $z=4$-$10$ ($29$ of which are spectroscopically confirmed with JWST NIRSpec), securing high signal-to-noise ratios of $10$-$135$ with deep JWST NIRCam $1$-$5μ$m images obtained by the CEERS survey. We derive morphologies of our high-$z$ galaxies, carefully evaluating the systematics of SB profile measurements with Monte Carlo simulations as well as the impacts of a) AGNs, b) multiple clumps including galaxy mergers, c) spatial resolution differences with previous HST studies, and d) strong emission lines, e.g., H$α$ and [OIII], on optical morphologies with medium-band F410M images. Conducting Sérsic profile fitting to our high-$z$ galaxy SBs with GALFIT, we obtain the effective radii of optical $r_{\rm e, opt}$ and UV $r_{\rm e, UV}$ wavelengths ranging $r_{\rm e, opt}=0.05$-$1.6$ kpc and $r_{\rm e, UV}=0.03$-$1.7$ kpc that are consistent with previous results within large scatters in the size luminosity relations. However, we find the effective radius ratio, $r_{\rm e, opt}/r_{\rm e, UV}$, is almost unity, $1.01^{+0.35}_{-0.22}$, over $z=4$-$10$ with no signatures of past inside-out star formation such found at $z\sim 0$-$2$. There are no spatial offsets exceeding $3σ$ between the optical and UV morphology centers in case of no mergers, indicative of major star-forming activity only found near a mass center of galaxies at $z\gtrsim 4$ probably experiencing the first phase of inside-out galaxy formation.

astro-ph.GA↗

arXiv2023

Efficient and Private Federated Trajectory Matching

Federated Trajectory Matching (FTM) is gaining increasing importance in big trajectory data analytics, supporting diverse applications such as public health, law enforcement, and emergency response. FTM retrieves trajectories that match with a query trajectory from a large-scale trajectory database, while safeguarding the privacy of trajectories in both the query and the database. A naive solution to FTM is to process the query through Secure Multi-Party Computation (SMC) across the entire database, which is inherently secure yet inevitably slow due to the massive secure operations. A promising acceleration strategy is to filter irrelevant trajectories from the database based on the query, thus reducing the SMC operations. However, a key challenge is how to publish the query in a way that both preserves privacy and enables efficient trajectory filtering. In this paper, we design GIST, a novel framework for efficient Federated Trajectory Matching. GIST is grounded in Geo-Indistinguishability, a privacy criterion dedicated to locations. It employs a new privacy mechanism for the query that facilitates efficient trajectory filtering. We theoretically prove the privacy guarantee of the mechanism and the accuracy of the filtering strategy of GIST. Extensive evaluations on five real datasets show that GIST is significantly faster and incurs up to 3 orders of magnitude lower communication cost than the state-of-the-arts.

cs.DB↗

arXiv2023

Latent Graph Inference with Limited Supervision

Latent graph inference (LGI) aims to jointly learn the underlying graph structure and node representations from data features. However, existing LGI methods commonly suffer from the issue of supervision starvation, where massive edge weights are learned without semantic supervision and do not contribute to the training loss. Consequently, these supervision-starved weights, which may determine the predictions of testing samples, cannot be semantically optimal, resulting in poor generalization. In this paper, we observe that this issue is actually caused by the graph sparsification operation, which severely destroys the important connections established between pivotal nodes and labeled ones. To address this, we propose to restore the corrupted affinities and replenish the missed supervision for better LGI. The key challenge then lies in identifying the critical nodes and recovering the corrupted affinities. We begin by defining the pivotal nodes as $k$-hop starved nodes, which can be identified based on a given adjacency matrix. Considering the high computational burden, we further present a more efficient alternative inspired by CUR matrix decomposition. Subsequently, we eliminate the starved nodes by reconstructing the destroyed connections. Extensive experiments on representative benchmarks demonstrate that reducing the starved nodes consistently improves the performance of state-of-the-art LGI methods, especially under extremely limited supervision (6.12% improvement on Pubmed with a labeling rate of only 0.3%).

cs.LG↗

arXiv2023

Pure Spectroscopic Constraints on UV Luminosity Functions and Cosmic Star Formation History From 25 Galaxies at $z_\mathrm{spec}=8.61-13.20$ Confirmed with JWST/NIRSpec

We present pure spectroscopic constraints on the UV luminosity functions and cosmic star formation rate (SFR) densities from 25 galaxies at $z_\mathrm{spec}=8.61-13.20$. By reducing the JWST/NIRSpec spectra taken in multiple programs of ERO, ERS, GO, and DDT with our analysis technique, we independently confirm 16 galaxies at $z_\mathrm{spec}=8.61-11.40$ including new redshift determinations, and a bright interloper at $z_\mathrm{spec}=4.91$ that was claimed as a photometric candidate at z~16. In conjunction with nine galaxies at redshifts up to $z_\mathrm{spec}=13.20$ in the literature, we make a sample of 25 spectroscopically-confirmed galaxies in total and carefully derive the best estimates and lower limits of the UV luminosity functions. These UV luminosity function constraints are consistent with the previous photometric estimates within the uncertainties and indicate mild redshift evolution towards z~12 showing tensions with some theoretical models of rapid evolution. With these spectroscopic constraints, we obtain firm lower limits of the cosmic SFR densities and spectroscopically confirm a high SFR density at z~12 beyond the constant star-formation efficiency models, which supports earlier claims from the photometric studies. While there are no spectroscopically-confirmed galaxies with very large stellar masses violating the $Λ$CDM model due to the removal of the bright interloper, we confirm star-forming galaxies at $z_\mathrm{spec}=11-13$ with stellar masses much higher than model predictions. Our results indicate possibilities of high star-formation efficiency (>5%), hidden AGN, top-heavy initial mass function (possibly with Pop-III), and large scatter/variance. Having these successful and unsuccessful spectroscopy results, we suggest observational strategies for efficiently removing low redshift interlopers for future JWST programs.

astro-ph.GA↗

arXiv2023

Learn to Optimize Denoising Scores for 3D Generation: A Unified and Improved Diffusion Prior on NeRF and 3D Gaussian Splatting

We propose a unified framework aimed at enhancing the diffusion priors for 3D generation tasks. Despite the critical importance of these tasks, existing methodologies often struggle to generate high-caliber results. We begin by examining the inherent limitations in previous diffusion priors. We identify a divergence between the diffusion priors and the training procedures of diffusion models that substantially impairs the quality of 3D generation. To address this issue, we propose a novel, unified framework that iteratively optimizes both the 3D model and the diffusion prior. Leveraging the different learnable parameters of the diffusion prior, our approach offers multiple configurations, affording various trade-offs between performance and implementation complexity. Notably, our experimental results demonstrate that our method markedly surpasses existing techniques, establishing new state-of-the-art in the realm of text-to-3D generation. Furthermore, our approach exhibits impressive performance on both NeRF and the newly introduced 3D Gaussian Splatting backbones. Additionally, our framework yields insightful contributions to the understanding of recent score distillation methods, such as the VSD and DDS loss.

cs.CV↗

arXiv2023

Background Clustering Pre-training for Few-shot Segmentation

Recent few-shot segmentation (FSS) methods introduce an extra pre-training stage before meta-training to obtain a stronger backbone, which has become a standard step in few-shot learning. Despite the effectiveness, current pre-training scheme suffers from the merged background problem: only base classes are labelled as foregrounds, making it hard to distinguish between novel classes and actual background. In this paper, we propose a new pre-training scheme for FSS via decoupling the novel classes from background, called Background Clustering Pre-Training (BCPT). Specifically, we adopt online clustering to the pixel embeddings of merged background to explore the underlying semantic structures, bridging the gap between pre-training and adaptation to novel classes. Given the clustering results, we further propose the background mining loss and leverage base classes to guide the clustering process, improving the quality and stability of clustering results. Experiments on PASCAL-5i and COCO-20i show that BCPT yields advanced performance. Code will be available.

cs.CV↗

arXiv2023

JWST Identification of Extremely Low C/N Galaxies with [N/O]$\gtrsim 0.5$ at $z\sim 6-10$ Evidencing the Early CNO-Cycle Enrichment and a Connection with Globular Cluster Formation

We present chemical abundance ratios of 70 star-forming galaxies at $z\sim4$-10 observed by the JWST/NIRSpec ERO, GLASS, and CEERS programs. Among the 70 galaxies, we have pinpointed 2 galaxies, CEERS_01019 at $z=8.68$ and GLASS_150008 at $z=6.23$, with extremely low C/N ([C/N]$\lesssim -1$), evidenced with CIII]$λλ$1907,1909, NIII]$λ$1750, and NIV]$λλ$1483,1486, which show high N/O ratios ([N/O]$\gtrsim 0.5$) comparable with the one of GN-z11 regardless of whether stellar or AGN radiation is assumed. Such low C/N and high N/O ratios found in CEERS_01019 and GLASS_150008 (additionally identified in GN-z11) are largely biased towards the equilibrium of the CNO cycle, suggesting that these 3 galaxies are enriched by metals processed by the CNO cycle. On the C/N vs. O/H plane, these 3 galaxies do not coincide with Galactic HII regions, normal star-forming galaxies, and nitrogen-loud quasars with asymptotic giant branch stars, but globular-cluster (GC) stars, indicating a connection with GC formation. We compare C/O and N/O of these 3 galaxies with those of theoretical models, and find that these 3 galaxies are explained by scenarios with dominant CNO-cycle materials, i.e. Wolf-Rayet stars, supermassive ($10^{3}-10^{5}\ M_{\odot}$) stars, and tidal disruption events, interestingly with a requirement of frequent direct collapses. For all the 70 galaxies, we present measurements of Ne/O, S/O, and Ar/O, together with C/O and N/O. We identify 4 galaxies with very low Ne/O, $\log(\rm Ne/O)<-1.0$, indicating abundant massive ($\gtrsim30\ M_\odot$) stars.

astro-ph.GA↗

arXiv2023

Supported Trust Region Optimization for Offline Reinforcement Learning

Offline reinforcement learning suffers from the out-of-distribution issue and extrapolation error. Most policy constraint methods regularize the density of the trained policy towards the behavior policy, which is too restrictive in most cases. We propose Supported Trust Region optimization (STR) which performs trust region policy optimization with the policy constrained within the support of the behavior policy, enjoying the less restrictive support constraint. We show that, when assuming no approximation and sampling error, STR guarantees strict policy improvement until convergence to the optimal support-constrained policy in the dataset. Further with both errors incorporated, STR still guarantees safe policy improvement for each step. Empirical results validate the theory of STR and demonstrate its state-of-the-art performance on MuJoCo locomotion domains and much more challenging AntMaze domains.

cs.LG↗

arXiv2023

Secure Inter-domain Routing and Forwarding via Verifiable Forwarding Commitments

The Internet inter-domain routing system is vulnerable. On the control plane, the de facto Border Gateway Protocol (BGP) does not have built-in mechanisms to authenticate routing announcements, so an adversary can announce virtually arbitrary paths to hijack network traffic; on the data plane, it is difficult to ensure that actual forwarding path complies with the control plane decisions. The community has proposed significant research to secure the routing system. Yet, existing secure BGP protocols (e.g., BGPsec) are not incrementally deployable, and existing path authorization protocols are not compatible with the current Internet routing infrastructure. In this paper, we propose FC-BGP, the first secure Internet inter-domain routing system that can simultaneously authenticate BGP announcements and validate data plane forwarding in an efficient and incrementally-deployable manner. FC-BGP is built upon a novel primitive, name Forwarding Commitment, to certify an AS's routing intent on its directly connected hops. We analyze the security benefits of FC-BGP in the Internet at different deployment rates. Further, we implement a prototype of FC-BGP and extensively evaluate it over a large-scale overlay network with 100 virtual machines deployed globally. The results demonstrate that FC-BGP saves roughly 55% of the overhead required to validate BGP announcements compared with BGPsec, and meanwhile FC-BGP introduces a small overhead for building a globally-consistent view on the desirable forwarding paths.

cs.NI↗

arXiv2023

A JWST/NIRSpec First Census of Broad-Line AGNs at z=4-7: Detection of 10 Faint AGNs with M_BH~10^6-10^8 M_sun and Their Host Galaxy Properties

We present a first statistical sample of faint type-1 AGNs at $z>4$ identified by JWST/NIRSpec deep spectroscopy. Among the 185 galaxies at $z_\mathrm{spec}=3.8-8.9$ confirmed with NIRSpec, our systematic search for broad-line emission reveals 10 type-1 AGNs at $z=4.015-6.936$ whose broad component is only seen in the permitted H$α$ line and not in the forbidden [OIII]$λ$5007 line that is detected with greater significance than H$α$. The broad H$α$ line widths of $\mathrm{FWHM}\simeq1000-6000\ \mathrm{km\ s^{-1}}$ suggest that the AGNs have low-mass black holes with $M_\mathrm{BH}\sim10^6-10^8\ M_\odot$, remarkably lower than those of low-luminosity quasars previously identified at $z>4$ with ground-based telescopes. JWST and HST high-resolution images reveal that the majority of them show extended morphologies indicating significant contribution to the total lights from their host galaxies, except for three compact objects two of which show red SEDs, probably in a transition phase from faint AGNs to low luminosity quasars. Careful AGN-host decomposition analyses show that their host's stellar masses are systematically lower than the local relation between the black hole mass and the stellar mass, implying a fast black hole growth consistent with predictions from theoretical simulations. A high fraction of the broad-line AGNs ($\sim5\%$), higher than $z\sim0$, indicates that a number density of such faint AGNs is higher than an extrapolation of the quasar luminosity function, implying a large population of AGNs including type 1 and type 2 in the early universe. Such faint AGNs contribute to cosmic reionization, while the total contribution is not large, up to $\sim50\%$ at $z\sim6$, because of their faint nature.

astro-ph.GA↗