arXiv subjects

Bo Zhang

Publications and source records attributed to Bo Zhang.

At least 523 records · Page 29Linked to original sources

arXiv2022

Radio continuum and OH line emission of high-z OH megamaser galaxies

We present the study of arcsecond scale radio continuum and OH line emission of a sample of known OH megamaser galaxies with $z \geq$ 0.15 using archival Very Large Array (VLA) data. And also the results of our pilot Five hundred meter aperture spherical radio telescope (FAST) observations of 12 of these OHM galaxies. The arcsecond-scale resolution images show that the OH emission is distributed in one compact structure and spatially associated with radio continuum emission. Furthermore, nearly all the fitted components are likely smaller than the beam size ($\sim$ 1.4"), which indicates that the broad OH line profiles of these sources originated from one masing region or that more components are distributed in sub-arcsec scales. The radio parameters, including brightness temperature, spectral index, and q-index, show no significant differences with the low-redshift OHM galaxies, which have significantly lower OH line luminosities. Because these parameters are indicators of the central power sources (AGN, starburst, or both), our results indicate that the presence of radio AGN in the nuclei may not be essential for the formation of OH emission. Over 1/3 of OHMs in this sample (6/17) show possible variable features likely caused by interstellar scintillation due to small angular sizes. We might underestimate this value because these sources are associated with this sample's highest OH line flux densities. Those with low OH line flux densities might need higher sensitivity observations to study the variabilities. These results support the compact nature of OH maser emission and a starburst origin for the OHMs in our selected sample.

astro-ph.GA↗

arXiv2022

Association between author metadata and acceptance: A feature-rich, matched observational study of a corpus of ICLR submissions between 2017-2022

Many recent studies have probed status bias in the peer-review process of academic journals and conferences. In this article, we investigated the association between author metadata and area chairs' final decisions (Accept/Reject) using our compiled database of 5,313 borderline submissions to the International Conference on Learning Representations (ICLR) from 2017 to 2022. We carefully defined elements in a cause-and-effect analysis, including the treatment and its timing, pre-treatment variables, potential outcomes and causal null hypothesis of interest, all in the context of study units being textual data and under Neyman and Rubin's potential outcomes (PO) framework. We found some weak evidence that author metadata was associated with articles' final decisions. We also found that, under an additional stability assumption, borderline articles from high-ranking institutions (top-30% or top-20%) were less favored by area chairs compared to their matched counterparts. The results were consistent in two different matched designs (odds ratio = 0.82 [95% CI: 0.67 to 1.00] in a first design and 0.83 [95% CI: 0.64 to 1.07] in a strengthened design). We discussed how to interpret these results in the context of multiple interactions between a study unit and different agents (reviewers and area chairs) in the peer-review system.

stat.AP↗

arXiv2022

Paint by Example: Exemplar-based Image Editing with Diffusion Models

Language-guided image editing has achieved great success recently. In this paper, for the first time, we investigate exemplar-guided image editing for more precise control. We achieve this goal by leveraging self-supervised training to disentangle and re-organize the source image and the exemplar. However, the naive approach will cause obvious fusing artifacts. We carefully analyze it and propose an information bottleneck and strong augmentations to avoid the trivial solution of directly copying and pasting the exemplar image. Meanwhile, to ensure the controllability of the editing process, we design an arbitrary shape mask for the exemplar image and leverage the classifier-free guidance to increase the similarity to the exemplar image. The whole framework involves a single forward of the diffusion model without any iterative optimization. We demonstrate that our method achieves an impressive performance and enables controllable editing on in-the-wild images with high fidelity.

cs.CV↗

arXiv2022

EAVN Astrometry toward the Extreme Outer Galaxy: Kinematic distance with the proper motion of G034.84-00.95

We aim to reveal the structure and kinematics of the Outer-Scutum-Centaurus (OSC) arm located on the far side of the Milky Way through very long baseline interferometry (VLBI) astrometry using KaVA, which is composed of KVN (Korean VLBI Network) and VERA (VLBI Exploration of Radio Astrometry). We report the proper motion of a 22 GHz H$_{2}$O maser source, which is associated with the star-forming region G034.84$-$00.95, to be ($μ_α \rm{cos}δ$, $μ_δ$) = ($-$1.61$\pm$0.18, $-$4.29$\pm$0.16) mas yr$^{-1}$ in equatorial coordinates (J2000). We estimate the 2D kinematic distance to the source to be 18.6$\pm$1.0 kpc, which is derived from the variance-weighted average of kinematic distances with LSR velocity and the Galactic-longitude component of the measured proper motion. Our result places the source in the OSC arm and implies that G034.84$-$00.95 is moving away from the Galactic plane with a vertical velocity of $-$38$\pm$16 km s$^{-1}$. Since the H I supershell GS033+06$-$49 is located at a kinematic distance roughly equal to that of G034.84$-$00.95, it is expected that gas circulation occurs between the outer Galactic disk around G034.84$-$00.95 with a Galactocentric distance of 12.8$^{+1.0}_{-0.9}$ kpc and halo. We evaluate possible origins of the fast vertical motion of G034.84$-$00.95, which are (1) supernova explosions and (2) cloud collisions with the Galactic disk. However, neither of the possibilities are matched with the results of VLBI astrometry as well as spatial distributions of H II regions and H I gas.

astro-ph.GA↗

arXiv2022

A Milliarcsecond-accurate Position for Sagittarius A*

The absolute position of Sgr A*, the compact radio source at the center of the Milky Way, had been uncertain by several tens of milliarcseconds. Here we report improved astrometric measurements of the absolute position and proper motion of Sgr A*. Three epochs of phase-referencing observations were conducted with the Very Long Baseline Array for Sgr A* at 22 and 43 GHz in 2019 and 2020. Using extragalactic radio sources with submilliarcsecond-accurate positions as reference, we determined the absolute position of Sgr A* at a reference epoch 2020.0 to be at $α$(J2000) = $17^{\rm h} 45^{\rm m}40.^{\rm s}032863~\pm~0.^{\rm s}000016$ and $δ$(J2000) = $-29^{\circ} 00^{\prime} 28.^{''}24260~\pm~0.^{''}00047$, with an updated proper motion $-3.152~\pm~0.011$ and $-5.586~\pm~0.006$ mas yr$^{-1}$ in the easterly and northerly directions, respectively.

astro-ph.GA↗

arXiv2022

Ternary Weight Networks

We present a memory and computation efficient ternary weight networks (TWNs) - with weights constrained to +1, 0 and -1. The Euclidian distance between full (float or double) precision weights and the ternary weights along with a scaling factor is minimized in training stage. Besides, a threshold-based ternary function is optimized to get an approximated solution which can be fast and easily computed. TWNs have shown better expressive abilities than binary precision counterparts. Meanwhile, TWNs achieve up to 16$\times$ model compression rate and need fewer multiplications compared with the float32 precision counterparts. Extensive experiments on MNIST, CIFAR-10, and ImageNet datasets show that the TWNs achieve much better result than the Binary-Weight-Networks (BWNs) and the classification performance on MNIST and CIFAR-10 is very close to the full precision networks. We also verify our method on object detection task and show that TWNs significantly outperforms BWN by more than 10\% mAP on PASCAL VOC dataset. The pytorch version of source code is available at: https://github.com/Thinklab-SJTU/twns.

cs.CV↗

arXiv2022

The Eclipsing Binaries from the LAMOST Medium-resolution Survey.III. A High-precision Empirical Stellar Mass Library

High-precision stellar mass and radius measured directly from binaries can effectively calibrate the stellar models. However, such a database containing full spectral types and large range of metallicity is still not fully established. A continuous effort of data collecting and analysis are requested to complete the database. In this work, we provide a catalog containing 184 binaries with independent atmospheric parameters and accurate masses and radii as the benchmark of stellar mass and radius. The catalog contains 56 new detached binaries from LAMOST Medium-resolution spectroscopic (MRS) survey and 128 detached eclipsing binaries compiled from previous studies. We obtain the orbital solutions of the new detached binaries with uncertainties of masses and radii smaller than 5%. These new samples densify the distribution of metallicity of the high-precision stellar mass library and add 9 hot stars with Teff>8000 K. Comparisons show that these samples well agree with the PARSEC isochrones in Teff-logg-mass-radius-luminosity space. We compare mass and radius estimates from isochrone and SED fitting, respectively, with those from the binary orbital solution. We find that the precision of the stellar-model dependent mass estimates is >10% and the precision of the radius estimates based on atmospheric parameters is >15%. These give a general view of the uncertainty of the usual approaches to estimate stellar mass and radius.

astro-ph.SR↗

arXiv2022

Instance-aware Model Ensemble With Distillation For Unsupervised Domain Adaptation

The linear ensemble based strategy, i.e., averaging ensemble, has been proposed to improve the performance in unsupervised domain adaptation tasks. However, a typical UDA task is usually challenged by dynamically changing factors, such as variable weather, views, and background in the unlabeled target domain. Most previous ensemble strategies ignore UDA's dynamic and uncontrollable challenge, facing limited feature representations and performance bottlenecks. To enhance the model, adaptability between domains and reduce the computational cost when deploying the ensemble model, we propose a novel framework, namely Instance aware Model Ensemble With Distillation, IMED, which fuses multiple UDA component models adaptively according to different instances and distills these components into a small model. The core idea of IMED is a dynamic instance aware ensemble strategy, where for each instance, a nonlinear fusion subnetwork is learned that fuses the extracted features and predicted labels of multiple component models. The nonlinear fusion method can help the ensemble model handle dynamically changing factors. After learning a large capacity ensemble model with good adaptability to different changing factors, we leverage the ensemble teacher model to guide the learning of a compact student model by knowledge distillation. Furthermore, we provide the theoretical analysis of the validity of IMED for UDA. Extensive experiments conducted on various UDA benchmark datasets, e.g., Office 31, Office Home, and VisDA 2017, show the superiority of the model based on IMED to the state of the art methods under the comparable computation cost.

cs.CV↗

arXiv2022

The statistical properties of early-type stars from LAMOST DR8

Massive binary stars play a crucial role in many astrophysical fields. Investigating the statistical properties of massive binary stars is essential to trace the formation of massive stars and constrain the evolution of stellar populations. However, no consensus has been achieved on the statistical properties of massive binary stars, mainly due to the lack of a large and homogeneous sample of spectroscopic observations. We study the intrinsic binary fraction $f_{\rm b}^{\rm in}$ and distributions of mass ratio $f(q)$ and orbital period $f(P)$ of early-type stars (comprised of O-, B-, and A-type stars) and investigate their dependences on effective temperature $T_{\rm eff}$, stellar metallicity [M/H], and the projection velocity $v\sin{i}$, based on the homogeneous spectroscopic sample from the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST) Data Release Eight (DR8). We found that $f_{\rm b}^{\rm in}$ increases with increasing $T_\mathrm{eff}$. The binary fraction is positively correlated with metallicity for spectra in the sample. Over all the $v\sin{i}$ values we considered, the $f_{\rm b}^{\rm in}$ have constant values of $\sim$50\%. It seems that the binary population is relatively evenly distributed over a wide range of $v\sin{i}$ values, while the whole sample shows that most of the stars are concentrated at low values of $v\sin{i}$ (probably from strong wind and magnetic braking of single massive stars) and at high values of $v\sin{i}$ (likely from the merging of binary stars). Stellar evolution and binary interaction may be partly responsible for this.There are no correlations found between $π$($γ$) and $T_{\rm eff}$, nor for $π$($γ$) and [M/H]. The uncertainties of the distribution decrease toward a larger sample size with higher observational cadence.

astro-ph.SR↗

arXiv2022

Efficient algorithms for building representative matched pairs with enhanced generalizability

Many recent efforts center on assessing the ability of real-world evidence (RWE) generated from non-randomized, observational data to produce results compatible with those from randomized controlled trials (RCTs). One noticeable endeavor is the RCT DUPLICATE initiative (Franklin et al., 2020, 2021). To better reconcile findings from an observational study and an RCT, or two observational studies based on different databases, it is desirable to eliminate differences between study populations. We outline an efficient, network-flow-based statistical matching algorithm that designs well-matched pairs from observational data that resemble the covariate distributions of a target population, for instance, the target-RCT-eligible population in the RCT DUPLICATE initiative studies or a generic population of scientific interest. We demonstrate the usefulness of the method by revisiting the inconsistency regarding a cardioprotective effect of the hormone replacement therapy (HRT) in the Women's Health Initiative (WHI) clinical trial and corresponding observational study. We found that the discrepancy between the trial and observational study persisted in a design that adjusted for study populations' cardiovascular risk profile, but seemed to disappear in a study design that further adjusted for the HRT initiation age and previous estrogen-plus-progestin use. The proposed method is integrated into the R package match2C.

stat.ME↗

arXiv2022

Multi-granularity Item-based Contrastive Recommendation

Contrastive learning (CL) has shown its power in recommendation. However, most CL-based recommendation models build their CL tasks merely focusing on the user's aspects, ignoring the rich diverse information in items. In this work, we propose a novel Multi-granularity item-based contrastive learning (MicRec) framework for the matching stage (i.e., candidate generation) in recommendation, which systematically introduces multi-aspect item-related information to representation learning with CL. Specifically, we build three item-based CL tasks as a set of plug-and-play auxiliary objectives to capture item correlations in feature, semantic and session levels. The feature-level item CL aims to learn the fine-grained feature-level item correlations via items and their augmentations. The semantic-level item CL focuses on the coarse-grained semantic correlations between semantically related items. The session-level item CL highlights the global behavioral correlations of items from users' sequential behaviors in all sessions. In experiments, we conduct both offline and online evaluations on real-world datasets, verifying the effectiveness and universality of three proposed CL tasks. Currently, MicRec has been deployed on a real-world recommender system, affecting millions of users. The source code will be released in the future.

cs.IR↗

arXiv2022

TESS Timings of 31 Hot Jupiters with Ephemeris Uncertainties

A precise transit ephemeris serves as the premise for follow-up exoplanet observations. We compare TESS Objects of Interest (TOI) transit timings of 262 hot Jupiters with the archival ephemeris and find 31 of them having TOI timing offsets, among which WASP-161b shows the most significant offset of -203.7$\pm$4.1 minutes. The median value of these offsets is 17.8 minutes, equivalent to 3.6 $σ$. We generate TESS timings in each sector for these 31 hot Jupiters, using a self-generated pipeline. The pipeline performs photometric measurements to TESS images and produces transit timings by fitting the light curves. We refine and update the previous ephemeris, based on these TESS timings (uncertainty $\sim$ 1 minute) and a long timing baseline ($\sim 10$ years). Our refined ephemeris gives the transit timing at a median precision of 0.82 minutes until 2025 and 1.21 minutes until 2030. We regard the timing offsets mainly originating from the underestimated ephemeris uncertainty. All the targets with timing offset larger than 10$σ$ present earlier timings than the prediction, which cannot be due to underestimated ephemeris uncertainty, apsidal precision, or R$ø$mer effect as those effects should be unsigned. For some particular targets, timing offsets are likely due to tidal dissipation. Our sample leads to the detection of period decaying candidates of WASP-161b and XO-3b reported previously.

astro-ph.EP↗

arXiv2022

SynGEC: Syntax-Enhanced Grammatical Error Correction with a Tailored GEC-Oriented Parser

This work proposes a syntax-enhanced grammatical error correction (GEC) approach named SynGEC that effectively incorporates dependency syntactic information into the encoder part of GEC models. The key challenge for this idea is that off-the-shelf parsers are unreliable when processing ungrammatical sentences. To confront this challenge, we propose to build a tailored GEC-oriented parser (GOPar) using parallel GEC training data as a pivot. First, we design an extended syntax representation scheme that allows us to represent both grammatical errors and syntax in a unified tree structure. Then, we obtain parse trees of the source incorrect sentences by projecting trees of the target correct sentences. Finally, we train GOPar with such projected trees. For GEC, we employ the graph convolution network to encode source-side syntactic information produced by GOPar, and fuse them with the outputs of the Transformer encoder. Experiments on mainstream English and Chinese GEC datasets show that our proposed SynGEC approach consistently and substantially outperforms strong baselines and achieves competitive performance. Our code and data are all publicly available at https://github.com/HillZhang1999/SynGEC.

cs.CL↗

arXiv2022

Integrated Decision and Control for High-Level Automated Vehicles by Mixed Policy Gradient and Its Experiment Verification

Self-evolution is indispensable to realize full autonomous driving. This paper presents a self-evolving decision-making system based on the Integrated Decision and Control (IDC), an advanced framework built on reinforcement learning (RL). First, an RL algorithm called constrained mixed policy gradient (CMPG) is proposed to consistently upgrade the driving policy of the IDC. It adapts the MPG under the penalty method so that it can solve constrained optimization problems using both the data and model. Second, an attention-based encoding (ABE) method is designed to tackle the state representation issue. It introduces an embedding network for feature extraction and a weighting network for feature fusion, fulfilling order-insensitive encoding and importance distinguishing of road users. Finally, by fusing CMPG and ABE, we develop the first data-driven decision and control system under the IDC architecture, and deploy the system on a fully-functional self-driving vehicle running in daily operation. Experiment results show that boosting by data, the system can achieve better driving ability over model-based methods. It also demonstrates safe, efficient and smart driving behavior in various complex scenes at a signalized intersection with real mixed traffic flow.

cs.RO↗

arXiv2022

3DFaceShop: Explicitly Controllable 3D-Aware Portrait Generation

In contrast to the traditional avatar creation pipeline which is a costly process, contemporary generative approaches directly learn the data distribution from photographs. While plenty of works extend unconditional generative models and achieve some levels of controllability, it is still challenging to ensure multi-view consistency, especially in large poses. In this work, we propose a network that generates 3D-aware portraits while being controllable according to semantic parameters regarding pose, identity, expression and illumination. Our network uses neural scene representation to model 3D-aware portraits, whose generation is guided by a parametric face model that supports explicit control. While the latent disentanglement can be further enhanced by contrasting images with partially different attributes, there still exists noticeable inconsistency in non-face areas, e.g., hair and background, when animating expressions. Wesolve this by proposing a volume blending strategy in which we form a composite output by blending dynamic and static areas, with two parts segmented from the jointly learned semantic field. Our method outperforms prior arts in extensive experiments, producing realistic portraits with vivid expression in natural lighting when viewed from free viewpoints. It also demonstrates generalization ability to real images as well as out-of-domain data, showing great promise in real applications.

cs.CV↗

arXiv2022

A Unified Mixture-View Framework for Unsupervised Representation Learning

Recent unsupervised contrastive representation learning follows a Single Instance Multi-view (SIM) paradigm where positive pairs are usually constructed with intra-image data augmentation. In this paper, we propose an effective approach called Beyond Single Instance Multi-view (BSIM). Specifically, we impose more accurate instance discrimination capability by measuring the joint similarity between two randomly sampled instances and their mixture, namely spurious-positive pairs. We believe that learning joint similarity helps to improve the performance when encoded features are distributed more evenly in the latent space. We apply it as an orthogonal improvement for unsupervised contrastive representation learning, including current outstanding methods SimCLR, MoCo, and BYOL. We evaluate our learned representations on many downstream benchmarks like linear classification on ImageNet-1k and PASCAL VOC 2007, object detection on MS COCO 2017 and VOC, etc. We obtain substantial gains with a large margin almost on all these tasks compared with prior arts.

cs.CV↗

arXiv2022

Sciences with Thai National Radio Telescope

This White Paper summarises potential key science topics to be achieved with Thai National Radio Telescope (TNRT). The commissioning phase has started in mid 2022. The key science topics consist of "Pulsars and Fast Radio Bursts (FRBs)", "Star Forming Regions (SFRs)", "Galaxy and Active Galactic Nuclei (AGNs)", "Evolved Stars", "Radio Emission of Chemically Peculiar (CP) Stars", and "Geodesy", covering a wide range of observing frequencies in L/C/X/Ku/K/Q/W-bands (1-115 GHz). As a single-dish instrument, TNRT is a perfect tool to explore time domain astronomy with its agile observing systems and flexible operation. Due to its ideal geographical location, TNRT will significantly enhance Very Long Baseline Interferometry (VLBI) arrays, such as East Asian VLBI Network (EAVN), Australia Long Baseline Array (LBA), European VLBI Network (EVN), in particular via providing a unique coverage of the sky resulting in a better complete "uv" coverage, improving synthesized-beam and imaging quality with reducing side-lobes. This document highlights key science topics achievable with TNRT in single-dish mode and in collaboration with VLBI arrays.

astro-ph.IM↗

arXiv2022

The Astrometric Animation of Water Masers towards the Mira Variable BX Cam

We report VLBI monitoring observations of the 22 GHz water (H$_{2}$O) masers around the Mira variable BX Cam, which were carried out as a part of the EAVN Synthesis of Stellar Maser Animations (ESTEMA) project. Data of 37 epochs in total were obtained from 2018 May to 2021 June with a time interval of 3-4 weeks, spanning approximately three stellar pulsation periods ($P= \sim$440 d). In particular, the dual-beam system equipped on the VERA stations was used to measure the kinematics and parallaxes of the H$_{2}$O maser features. The measured parallax, $π=1.79\pm 0.08$ mas, is consistent with $Gaia$ EDR3 and previously measured VLBI parallaxes within a 1-$σ$ error level. The position of the central star was estimated, based on both the $Gaia$ EDR3 data and the center position of the ring-like 43 GHz silicon-monoxide (SiO) maser distribution imaged with the KVN. The three-dimensional H$_{2}$O maser kinematics indicates that the circumstellar envelope is expanding at a velocity of $13\pm4$ km s$^{-1}$, while there are asymmetries in both the spatial and velocity distributions of the maser features. Furthermore, the H$_{2}$O maser animation achieved by our dense monitoring program manifests the propagation of shock waves in the circumstellar envelope of BX Cam.

astro-ph.SR↗