arXiv subjects

Yan Huang

Publications and source records attributed to Yan Huang.

At least 235 records · Page 13Linked to original sources

arXiv2021

FDAN: Flow-guided Deformable Alignment Network for Video Super-Resolution

Most Video Super-Resolution (VSR) methods enhance a video reference frame by aligning its neighboring frames and mining information on these frames. Recently, deformable alignment has drawn extensive attention in VSR community for its remarkable performance, which can adaptively align neighboring frames with the reference one. However, we experimentally find that deformable alignment methods still suffer from fast motion due to locally loss-driven offset prediction and lack explicit motion constraints. Hence, we propose a Matching-based Flow Estimation (MFE) module to conduct global semantic feature matching and estimate optical flow as coarse offset for each location. And a Flow-guided Deformable Module (FDM) is proposed to integrate optical flow into deformable convolution. The FDM uses the optical flow to warp the neighboring frames at first. And then, the warped neighboring frames and the reference one are used to predict a set of fine offsets for each coarse offset. In general, we propose an end-to-end deep network called Flow-guided Deformable Alignment Network (FDAN), which reaches the state-of-the-art performance on two benchmark datasets while is still competitive in computation and memory consumption.

cs.CV↗

arXiv2021

Search for Light Dark Matter-Electron Scatterings in the PandaX-II Experiment

We report constraints on light dark matter through its interactions with shell electrons in the PandaX-II liquid xenon detector with a total 46.9 tonne$\cdot$day exposure. To effectively search for these very low energy electron recoils, ionization-only signals are selected from the data. 1821 candidates are identified within ionization signal range between 50 to 75 photoelectrons, corresponding to a mean electronic recoil energy from 0.08 to 0.15 keV. The 90% C.L. exclusion limit on the scattering cross section between the dark matter and electron is calculated based on Poisson statistics. Under the assumption of point interaction, we provide the world's most stringent limit within the dark matter mass range from 15 to 30 $\rm MeV/c^2$, with the corresponding cross section from $2.5\times10^{-37}$ to $3.1\times10^{-38}$ cm$^2$.

hep-ex↗

arXiv2021

Kilonova Emission From Black Hole-Neutron Star Mergers. II. Luminosity Function and Implications for Target-of-opportunity Observations of Gravitational-wave Triggers and Blind Searches

We present detailed simulations of black hole-neutron star (BH-NS) mergers kilonova and gamma-ray burst (GRB) afterglow and kilonova luminosity function, and discuss the detectability of electromagnetic (EM) counterpart in connection with gravitational wave (GW) detections, GW-triggered target-of-opportunity observations, and time-domain blind searches. The predicted absolute magnitude of the BH-NS kilonovae at $0.5\,{\rm days}$ after the merger falls in $[-10,-15.5]$. The simulated luminosity function contains the potential viewing-angle distribution information of the anisotropic kilonova emission. We simulate the GW detection rates, detectable distances and signal duration, for the future networks of 2nd/2.5th/3rd-generation GW detectors. BH-NSs tend to produce brighter kilonovae and afterglows if the BH has a higher aligned-spin, and a less massive NS with a stiffer EoS. The detectability of kilonova is especially sensitive to the BH spin. If BHs typically have low spins, the BH-NS EM counterparts are hard to discover. For the 2nd generation GW detector networks, a limiting magnitude of $m_{\rm limit}\sim23-24\,{\rm mag}$ is required to detect the kilonovae even if BH high spin is assumed. Thus, a plausible explanation for the lack of BH-NS associated kilonova detection during LIGO/Virgo O3 is that either there is no EM counterpart (plunging events), or the current follow-ups are too shallow. These observations still have the chance to detect the on-axis jet afterglow associated with an sGRB or an orphan afterglow. Follow-up observations can detect possible associated sGRB afterglows, from which kilonova signatures may be studied. For time-domain observations, a high-cadence search in redder filters is recommended to detect more BH-NS associated kilonovae and afterglows.

astro-ph.HE↗

arXiv2021

Rethinking the Heatmap Regression for Bottom-up Human Pose Estimation

Heatmap regression has become the most prevalent choice for nowadays human pose estimation methods. The ground-truth heatmaps are usually constructed via covering all skeletal keypoints by 2D gaussian kernels. The standard deviations of these kernels are fixed. However, for bottom-up methods, which need to handle a large variance of human scales and labeling ambiguities, the current practice seems unreasonable. To better cope with these problems, we propose the scale-adaptive heatmap regression (SAHR) method, which can adaptively adjust the standard deviation for each keypoint. In this way, SAHR is more tolerant of various human scales and labeling ambiguities. However, SAHR may aggravate the imbalance between fore-background samples, which potentially hurts the improvement of SAHR. Thus, we further introduce the weight-adaptive heatmap regression (WAHR) to help balance the fore-background samples. Extensive experiments show that SAHR together with WAHR largely improves the accuracy of bottom-up human pose estimation. As a result, we finally outperform the state-of-the-art model by +1.5AP and achieve 72.0AP on COCO test-dev2017, which is com-arable with the performances of most top-down methods. Source codes are available at https://github.com/greatlog/SWAHR-HumanPose.

cs.CV↗

arXiv2021

Determination of responses of liquid xenon to low energy electron and nuclear recoils using the PandaX-II detector

We report a systematic determination of the responses of PandaX-II, a dual phase xenon time projection chamber detector, to low energy recoils. The electron recoil (ER) and nuclear recoil (NR) responses are calibrated, respectively, with injected tritiated methane or $^{220}$Rn source, and with $^{241}$Am-Be neutron source, within an energy range from $1-25$ keV (ER) and $4-80$ keV (NR), under the two drift fields of 400 and 317 V/cm. An empirical model is used to fit the light yield and charge yield for both types of recoils. The best fit models can well describe the calibration data. The systematic uncertainties of the fitted models are obtained via statistical comparison against the data.

physics.ins-det↗

arXiv2021

FWB-Net:Front White Balance Network for Color Shift Correction in Single Image Dehazing via Atmospheric Light Estimation

In recent years, single image dehazing deep models based on Atmospheric Scattering Model (ASM) have achieved remarkable results. But the dehazing outputs of those models suffer from color shift. Analyzing the ASM model shows that the atmospheric light factor (ALF) is set as a scalar which indicates ALF is constant for whole image. However, for images taken in real-world, the illumination is not uniformly distributed over whole image which brings model mismatch and possibly results in color shift of the deep models using ASM. Bearing this in mind, in this study, first, a new non-homogeneous atmospheric scattering model (NH-ASM) is proposed for improving image modeling of hazy images taken under complex illumination conditions. Second, a new U-Net based front white balance module (FWB-Module) is dedicatedly designed to correct color shift before generating dehazing result via atmospheric light estimation. Third, a new FWB loss is innovatively developed for training FWB-Module, which imposes penalty on color shift. In the end, based on NH-ASM and front white balance technology, an end-to-end CNN-based color-shift-restraining dehazing network is developed, termed as FWB-Net. Experimental results demonstrate the effectiveness and superiority of our proposed FWB-Net for dehazing on both synthetic and real-world images.

cs.CV↗

arXiv2021

Internal Calibration of the PandaX-II Detector with Radon Gaseous Sources

We have developed a low-energy electron recoil (ER) calibration method with $^{220}$Rn for the PandaX-II detector. $^{220}$Rn, emanated from natural thorium compounds, was fed into the detector through the xenon purification system. From 2017 to 2019, we performed three dedicated calibration campaigns with different radon sources. We studied the detector response to $α$, $β$, and $γ$ particles with focus on low energy ER events. During the runs in 2017 and 2018, the amount of radioactivity of $^{222}$Rn were on the order of 1\% of that of $^{220}$Rn and thorium particulate contamination was negligible, especially in 2018. We also measured the background contribution from $^{214}$Pb for the first time in PandaX-II with the help from a $^{222}$Rn injection. Calibration strategy with $^{220}$Rn and $^{222}$Rn will be implemented in the upcoming PandaX-4T experiment and can be useful for other xenon-based detectors as well.

physics.ins-det↗

arXiv2021

Results of Dark Matter Search using the Full PandaX-II Exposure

We report the dark matter search results obtained using the full 132 ton$\cdot$day exposure of the PandaX-II experiment, including all data from March 2016 to August 2018. No significant excess of events is identified above the expected background. Upper limits are set on the spin-independent dark matter-nucleon interactions. The lowest 90% confidence level exclusion on the spin-independent cross section is $2.2\times 10^{-46}$ cm$^2$ at a WIMP mass of 30 GeV/$c^2$.

astro-ph.CO↗

arXiv2020

Efficient Human Pose Estimation by Learning Deeply Aggregated Representations

In this paper, we propose an efficient human pose estimation network (DANet) by learning deeply aggregated representations. Most existing models explore multi-scale information mainly from features with different spatial sizes. Powerful multi-scale representations usually rely on the cascaded pyramid framework. This framework largely boosts the performance but in the meanwhile makes networks very deep and complex. Instead, we focus on exploiting multi-scale information from layers with different receptive-field sizes and then making full of use this information by improving the fusion method. Specifically, we propose an orthogonal attention block (OAB) and a second-order fusion unit (SFU). The OAB learns multi-scale information from different layers and enhances them by encouraging them to be diverse. The SFU adaptively selects and fuses diverse multi-scale information and suppress the redundant ones. This could maximize the effective information in final fused representations. With the help of OAB and SFU, our single pyramid network may be able to generate deeply aggregated representations that contain even richer multi-scale information and have a larger representing capacity than that of cascaded networks. Thus, our networks could achieve comparable or even better accuracy with much smaller model complexity. Specifically, our \mbox{DANet-72} achieves $70.5$ in AP score on COCO test-dev set with only $1.0G$ FLOPs. Its speed on a CPU platform achieves $58$ Persons-Per-Second~(PPS).

cs.CV↗

arXiv2020

Observation of Magnetic Droplets in Magnetic Tunnel Junctions

Magnetic droplets, a class of highly non-linear magnetodynamical solitons, can be nucleated and stabilized in nanocontact spin-torque nano-oscillators where they greatly increase the microwave output power. Here, we experimentally demonstrate magnetic droplets in magnetic tunnel junctions (MTJs). The droplet nucleation is accompanied by a power increase of over 300 times compared to its ferromagnetic resonance modes. The nucleation and stabilization of droplets are ascribed to the double-CoFeB free layer structure in the all-perpendicular MTJ which provides a low Zhang-Li torque and a high pinning field. Our results enable better electrical sensitivity in the fundamental studies of droplets and show that the droplets can be utilized in MTJ-based applications.

physics.app-ph↗

arXiv2020

Unfolding the Alternating Optimization for Blind Super Resolution

Previous methods decompose blind super resolution (SR) problem into two sequential steps: \textit{i}) estimating blur kernel from given low-resolution (LR) image and \textit{ii}) restoring SR image based on estimated kernel. This two-step solution involves two independently trained models, which may not be well compatible with each other. Small estimation error of the first step could cause severe performance drop of the second one. While on the other hand, the first step can only utilize limited information from LR image, which makes it difficult to predict highly accurate blur kernel. Towards these issues, instead of considering these two steps separately, we adopt an alternating optimization algorithm, which can estimate blur kernel and restore SR image in a single model. Specifically, we design two convolutional neural modules, namely \textit{Restorer} and \textit{Estimator}. \textit{Restorer} restores SR image based on predicted kernel, and \textit{Estimator} estimates blur kernel with the help of restored SR image. We alternate these two modules repeatedly and unfold this process to form an end-to-end trainable network. In this way, \textit{Estimator} utilizes information from both LR and SR images, which makes the estimation of blur kernel easier. More importantly, \textit{Restorer} is trained with the kernel estimated by \textit{Estimator}, instead of ground-truth kernel, thus \textit{Restorer} could be more tolerant to the estimation error of \textit{Estimator}. Extensive experiments on synthetic datasets and real-world images show that our model can largely outperform state-of-the-art methods and produce more visually favorable results at much higher speed. The source code is available at https://github.com/greatlog/DAN.git.

cs.CV↗

arXiv2020

Collaborative City Digital Twin For Covid-19 Pandemic: A Federated Learning Solution

In this work, we propose a collaborative city digital twin based on FL, a novel paradigm that allowing multiple city DT to share the local strategy and status in a timely manner. In particular, an FL central server manages the local updates of multiple collaborators (city DT), provides a global model which is trained in multiple iterations at different city DT systems, until the model gains the correlations between various response plan and infection trend. That means, a collaborative city DT paradigm based on FL techniques can obtain knowledge and patterns from multiple DTs, and eventually establish a `global view' for city crisis management. Meanwhile, it also helps to improve each city digital twin selves by consolidating other DT's respective data without violating privacy rules. To validate the proposed solution, we take COVID-19 pandemic as a case study. The experimental results on the real dataset with various response plan validate our proposed solution and demonstrate the superior performance.

cs.LG↗

arXiv2020

Algorithmic Transparency with Strategic Users

Should firms that apply machine learning algorithms in their decision-making make their algorithms transparent to the users they affect? Despite growing calls for algorithmic transparency, most firms have kept their algorithms opaque, citing potential gaming by users that may negatively affect the algorithm's predictive power. We develop an analytical model to compare firm and user surplus with and without algorithmic transparency in the presence of strategic users and present novel insights. We identify a broad set of conditions under which making the algorithm transparent benefits the firm. We show that, in some cases, even the predictive power of machine learning algorithms may increase if the firm makes them transparent. By contrast, users may not always be better off under algorithmic transparency. The results hold even when the predictive power of the opaque algorithm comes largely from correlational features and the cost for users to improve on them is close to zero. Overall, our results show that firms should not view manipulation by users as bad. Rather, they should use algorithmic transparency as a lever to motivate users to invest in more desirable features.

cs.GT↗

arXiv2020

Recurrent Deconvolutional Generative Adversarial Networks with Application to Text Guided Video Generation

This paper proposes a novel model for video generation and especially makes the attempt to deal with the problem of video generation from text descriptions, i.e., synthesizing realistic videos conditioned on given texts. Existing video generation methods cannot be easily adapted to handle this task well, due to the frame discontinuity issue and their text-free generation schemes. To address these problems, we propose a recurrent deconvolutional generative adversarial network (RD-GAN), which includes a recurrent deconvolutional network (RDN) as the generator and a 3D convolutional neural network (3D-CNN) as the discriminator. The RDN is a deconvolutional version of conventional recurrent neural network, which can well model the long-range temporal dependency of generated video frames and make good use of conditional information. The proposed model can be jointly trained by pushing the RDN to generate realistic videos so that the 3D-CNN cannot distinguish them from real ones. We apply the proposed RD-GAN to a series of tasks including conventional video generation, conditional video generation, video prediction and video classification, and demonstrate its effectiveness by achieving well performance.

cs.CV↗

arXiv2020

Crowd, Lending, Machine, and Bias

Big data and machine learning (ML) algorithms are key drivers of many fintech innovations. While it may be obvious that replacing humans with machine would increase efficiency, it is not clear whether and where machines can make better decisions than humans. We answer this question in the context of crowd lending, where decisions are traditionally made by a crowd of investors. Using data from Prosper.com, we show that a reasonably sophisticated ML algorithm predicts listing default probability more accurately than crowd investors. The dominance of the machine over the crowd is more pronounced for highly risky listings. We then use the machine to make investment decisions, and find that the machine benefits not only the lenders but also the borrowers. When machine prediction is used to select loans, it leads to a higher rate of return for investors and more funding opportunities for borrowers with few alternative funding options. We also find suggestive evidence that the machine is biased in gender and race even when it does not use gender and race information as input. We propose a general and effective "debasing" method that can be applied to any prediction focused ML applications, and demonstrate its use in our context. We show that the debiased ML algorithm, which suffers from lower prediction accuracy, still leads to better investment decisions compared with the crowd. These results indicate that ML can help crowd lending platforms better fulfill the promise of providing access to financial resources to otherwise underserved individuals and ensure fairness in the allocation of these resources.

q-fin.GN↗

arXiv2020

Kilonova Emission From Black Hole-Neutron Star Mergers. I. Viewing-Angle-Dependent Lightcurves

In this paper, we present a numerical method to study the predicted lightcurves as a function of viewing angle. We extrapolate the fitting formulae for the mass and velocity of tidal dynamical ejecta across a wide range of mass ratio validated with 66 simulations and use them in the calculations of the kilonova lightcurves. The calculated peak luminosity of a BH-NS merger kilonova is typically about a few times $10^{41}\ {\rm erg\ s^{-1}}$, which is always $\lesssim4.5\times10^{41}\ {\rm erg\ s^{-1}}$. This corresponds to the AB absolute magnitudes fainter than $\sim -15\ {\rm mag}$ in optical and $\sim -16\ {\rm mag}$ in infrared. Since the projected photosphere area of the dynamical ejecta is much larger than that of the disk wind outflows, the dynamical ejecta usually contribute to the majority of the kilonova emission from BH-NS mergers. The fitted blackbody temperature and the shape of the observed multi-band lightcurves are insensitive to the line of sight. The peak time of the observed multi-band lightcurves, affected by the light propagation effect, is related to the relative motion direction between the dynamical ejecta and the observer. The observed luminosity varies with the projected photosphere area determined by the viewing angles. However, the predicted peak luminosity only varies by a factor of $\sim (2 - 3)$ (or by $\sim1\ {\rm mag}$) for different viewing angles. When the short-duration gamma-ray burst afterglow is taken into account, for an on-axis geometry, the kilonova emission is usually outshone by the afterglow emission and can be only observed in the redder bands, especially in the $K$-band at late times. Compared with GW170817/AT2017gfo, the BH-NS merger kilonovae are optically dim but possibly infrared bright. At the same epoch after the merger, the blackbody fitting temperature of the BH-NS merger kilonovae is lower than that of GW170817/AT2017gfo.

astro-ph.HE↗

arXiv2020

L-Vector: Neural Label Embedding for Domain Adaptation

We propose a novel neural label embedding (NLE) scheme for the domain adaptation of a deep neural network (DNN) acoustic model with unpaired data samples from source and target domains. With NLE method, we distill the knowledge from a powerful source-domain DNN into a dictionary of label embeddings, or l-vectors, one for each senone class. Each l-vector is a representation of the senone-specific output distributions of the source-domain DNN and is learned to minimize the average L2, Kullback-Leibler (KL) or symmetric KL distance to the output vectors with the same label through simple averaging or standard back-propagation. During adaptation, the l-vectors serve as the soft targets to train the target-domain model with cross-entropy loss. Without parallel data constraint as in the teacher-student learning, NLE is specially suited for the situation where the paired target-domain data cannot be simulated from the source-domain data. We adapt a 6400 hours multi-conditional US English acoustic model to each of the 9 accented English (80 to 830 hours) and kids' speech (80 hours). NLE achieves up to 14.1% relative word error rate reduction over direct re-training with one-hot labels.

eess.AS↗

arXiv2020

Learning Goal-oriented Dialogue Policy with Opposite Agent Awareness

Most existing approaches for goal-oriented dialogue policy learning used reinforcement learning, which focuses on the target agent policy and simply treat the opposite agent policy as part of the environment. While in real-world scenarios, the behavior of an opposite agent often exhibits certain patterns or underlies hidden policies, which can be inferred and utilized by the target agent to facilitate its own decision making. This strategy is common in human mental simulation by first imaging a specific action and the probable results before really acting it. We therefore propose an opposite behavior aware framework for policy learning in goal-oriented dialogues. We estimate the opposite agent's policy from its behavior and use this estimation to improve the target agent by regarding it as part of the target policy. We evaluate our model on both cooperative and competitive dialogue tasks, showing superior performance over state-of-the-art baselines.

cs.CL↗