arXiv subjects

Jun Zhang

Publications and source records attributed to Jun Zhang.

At least 739 records · Page 41Linked to original sources

arXiv2022

When Optimal Transport Meets Information Geometry

Information geometry and optimal transport are two distinct geometric frameworks for modeling families of probability measures. During the recent years, there has been a surge of research endeavors that cut across these two areas and explore their links and interactions. This paper is intended to provide an (incomplete) survey of these works, including entropy-regularized transport, divergence functions arising from $c$-duality, density manifolds and transport information geometry, the para-Kähler and Kähler geometries underlying optimal transport and the regularity theory for its solutions. Some outstanding questions that would be of interest to audience of both these two disciplines are posed. Our piece also serves as an introduction to the Special Issue on Optimal Transport of the journal Information Geometry.

math.OC↗

arXiv2022

PSP: Million-level Protein Sequence Dataset for Protein Structure Prediction

Proteins are essential component of human life and their structures are important for function and mechanism analysis. Recent work has shown the potential of AI-driven methods for protein structure prediction. However, the development of new models is restricted by the lack of dataset and benchmark training procedure. To the best of our knowledge, the existing open source datasets are far less to satisfy the needs of modern protein sequence-structure related research. To solve this problem, we present the first million-level protein structure prediction dataset with high coverage and diversity, named as PSP. This dataset consists of 570k true structure sequences (10TB) and 745k complementary distillation sequences (15TB). We provide in addition the benchmark training procedure for SOTA protein structure prediction model on this dataset. We validate the utility of this dataset for training by participating CAMEO contest in which our model won the first place. We hope our PSP dataset together with the training benchmark can enable a broader community of AI/biology researchers for AI-driven protein related research.

q-bio.BM↗

arXiv2022

A Manifold-based Airfoil Geometric-feature Extraction and Discrepant Data Fusion Learning Method

Geometrical shape of airfoils, together with the corresponding flight conditions, are crucial factors for aerodynamic performances prediction. The obtained airfoils geometrical features in most existing approaches (e.g., geometrical parameters extraction, polynomial description and deep learning) are in Euclidean space. State-of-the-art studies showed that curves or surfaces of an airfoil formed a manifold in Riemannian space. Therefore, the features extracted by existing methods are not sufficient to reflect the geometric-features of airfoils. Meanwhile, flight conditions and geometric features are greatly discrepant with different types, the relevant knowledge of the influence of these two factors that on final aerodynamic performances predictions must be evaluated and learned to improve prediction accuracy. Motivated by the advantages of manifold theory and multi-task learning, we propose a manifold-based airfoil geometric-feature extraction and discrepant data fusion learning method (MDF) to extract geometric-features of airfoils in Riemannian space (we call them manifold-features) and further fuse the manifold-features with flight conditions to predict aerodynamic performances. Experimental results show that our method could extract geometric-features of airfoils more accurately compared with existing methods, that the average MSE of re-built airfoils is reduced by 56.33%, and while keeping the same predicted accuracy level of CL, the MSE of CD predicted by MDF is further reduced by 35.37%.

cs.LG↗

arXiv2022

Model-Driven Deep Learning-Based MIMO-OFDM Detector: Design, Simulation, and Experimental Results

Multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM), a fundamental transmission scheme, promises high throughput and robustness against multipath fading. However, these benefits rely on the efficient detection strategy at the receiver and come at the expense of the extra bandwidth consumed by the cyclic prefix (CP). We use the iterative orthogonal approximate message passing (OAMP) algorithm in this paper as the prototype of the detector because of its remarkable potential for interference suppression. However, OAMP is computationally expensive for the matrix inversion per iteration. We replace the matrix inversion with the conjugate gradient (CG) method to reduce the complexity of OAMP. We further unfold the CG-based OAMP algorithm into a network and tune the critical parameters through deep learning (DL) to enhance detection performance. Simulation results and complexity analysis show that the proposed scheme has significant gain over other iterative detection methods and exhibits comparable performance to the state-of-the-art DL-based detector at a reduced computational cost. Furthermore, we design a highly efficient CP-free MIMO-OFDM receiver architecture to remove the CP overhead. This architecture first eliminates the intersymbol interference by buffering the previously recovered data and then detects the signal using the proposed detector. Numerical experiments demonstrate that the designed receiver offers a higher spectral efficiency than traditional receivers. Finally, over-the-air tests verify the effectiveness and robustness of the proposed scheme in realistic environments.

cs.IT↗

arXiv2022

Resource-Constrained Edge AI with Early Exit Prediction

By leveraging the data sample diversity, the early-exit network recently emerges as a prominent neural network architecture to accelerate the deep learning inference process. However, intermediate classifiers of the early exits introduce additional computation overhead, which is unfavorable for resource-constrained edge artificial intelligence (AI). In this paper, we propose an early exit prediction mechanism to reduce the on-device computation overhead in a device-edge co-inference system supported by early-exit networks. Specifically, we design a low-complexity module, namely the Exit Predictor, to guide some distinctly "hard" samples to bypass the computation of the early exits. Besides, considering the varying communication bandwidth, we extend the early exit prediction mechanism for latency-aware edge inference, which adapts the prediction thresholds of the Exit Predictor and the confidence thresholds of the early-exit network via a few simple regression models. Extensive experiment results demonstrate the effectiveness of the Exit Predictor in achieving a better tradeoff between accuracy and on-device computation overhead for early-exit networks. Besides, compared with the baseline methods, the proposed method for latency-aware edge inference attains higher inference accuracy under different bandwidth conditions.

cs.LG↗

arXiv2022

ET White Paper: To Find the First Earth 2.0

We propose to develop a wide-field and ultra-high-precision photometric survey mission, temporarily named "Earth 2.0 (ET)". This mission is designed to measure, for the first time, the occurrence rate and the orbital distributions of Earth-sized planets. ET consists of seven 30cm telescopes, to be launched to the Earth-Sun's L2 point. Six of these are transit telescopes with a field of view of 500 square degrees. Staring in the direction that encompasses the original Kepler field for four continuous years, this monitoring will return tens of thousands of transiting planets, including the elusive Earth twins orbiting solar-type stars. The seventh telescope is a 30cm microlensing telescope that will monitor an area of 4 square degrees toward the galactic bulge. This, combined with simultaneous ground-based KMTNet observations, will measure masses for hundreds of long-period and free-floating planets. Together, the transit and the microlensing telescopes will revolutionize our understandings of terrestrial planets across a large swath of orbital distances and free space. In addition, the survey data will also facilitate studies in the fields of asteroseismology, Galactic archeology, time-domain sciences, and black holes in binaries.

astro-ph.IM↗

arXiv2022

Federated Learning with GAN-based Data Synthesis for Non-IID Clients

Federated learning (FL) has recently emerged as a popular privacy-preserving collaborative learning paradigm. However, it suffers from the non-independent and identically distributed (non-IID) data among clients. In this paper, we propose a novel framework, named Synthetic Data Aided Federated Learning (SDA-FL), to resolve this non-IID challenge by sharing synthetic data. Specifically, each client pretrains a local generative adversarial network (GAN) to generate differentially private synthetic data, which are uploaded to the parameter server (PS) to construct a global shared synthetic dataset. To generate confident pseudo labels for the synthetic dataset, we also propose an iterative pseudo labeling mechanism performed by the PS. A combination of the local private dataset and synthetic dataset with confident pseudo labels leads to nearly identical data distributions among clients, which improves the consistency among local models and benefits the global aggregation. Extensive experiments evidence that the proposed framework outperforms the baseline methods by a large margin in several benchmark datasets under both the supervised and semi-supervised settings.

cs.LG↗

arXiv2022

Holographic entanglement entropy of the double Wick rotated BTZ black hole

In this paper, we analyze the holographic covariant entanglement entropy in the double Wick rotated version of a rotating BTZ black hole (3 dimensional Kerr-AdS solution), where the periodicity of Euclidean time and spatial direction are changed. The dual field theory has negative energy in the Lorentzian signature. The holographic entanglement entropy agrees with its CFT counterpart, which is obtained by a conformal transformation of the correlation functions of twisted operators.

hep-th↗

arXiv2022

Miniaturizing Color-Sensitive Photodetectors via Hybrid Nanoantennas towards Sub-micron Dimensions

Digital camera sensors utilize color filters on photodiodes to achieve color selectivity. As color filters and photosensitive silicon layers are separate elements, these sensors suffer from optical cross-talk, which sets limits to the minimum pixel size. In this paper, we report hybrid silicon-aluminum nanostructures in the extreme limit of zero distance between color filters and sensors. This design could essentially achieve sub micron pixel dimensions and minimize the optical cross-talk originated from tilt illuminations. The designed hybrid silicon-aluminum nanostructure has dual functionalities. Crucially, it supports a hybrid Mie-plasmon resonance of magnetic dipole to achieve the color-selective light absorption, generating electron hole pairs. Simultaneously, the silicon-aluminum interface forms a Schottky barrier for charge separation and photodetection. This design could potentially replace the traditional dye based filters for camera sensors at ultra-high pixel densities with advanced functionalities in sensing polarization and directionality, as well as UV selectivity via interband plasmons of silicon.

physics.optics↗

arXiv2022

Shear Measurement with Poorly Resolved Images

Weak lensing studies typically require excellent seeing conditions for the purpose of maximizing the number density of well-resolved galaxy images. It is interesting to ask to what extent the seeing size limits the usefulness of the astronomical images in weak lensing. In this work, we study this issue with the data of the DECam Legacy Survey (DECaLS), which is a part of the target selection program for the Dark Energy Spectroscopic Instrument (DESI). Using the Fourier Quad shear measurement pipeline, we demonstrate that images with relatively poor seeing conditions (around 1.5 arcsec) can still yield accurate shear estimators. We do not find any correlation between systematic shear error and the image resolution.

astro-ph.CO↗

arXiv2022

Tolerance For the Pixelation Effect in Shear Measurement

Images taken by space telescopes typically have a superb spatial resolution, but a relatively poor sampling rate due to the finite CCD pixel size. Beyond the Nyquist limit, it becomes uncertain how much the pixelation effect may affect the accuracy of galaxy shape measurement. It is timely to study this issue given that a number of space-based large-scale weak lensing surveys are planned. Using the Fourier_Quad method, we quantify the shear recovery error as a function of the sampling factor Q, i.e., the ratio between the FWHM of the point-spread-function (PSF) and the pixel size of the CCD, for different PSFs and galaxies of different sizes and noise levels. We show that sub-percent-level accuracy in shear recovery is achievable with single-exposure images for $Q\lesssim 2$. The conclusion holds for galaxies much smaller than the PSF, and those with a significant level of noise.

astro-ph.CO↗

arXiv2022

PyramidCLIP: Hierarchical Feature Alignment for Vision-language Model Pretraining

Large-scale vision-language pre-training has achieved promising results on downstream tasks. Existing methods highly rely on the assumption that the image-text pairs crawled from the Internet are in perfect one-to-one correspondence. However, in real scenarios, this assumption can be difficult to hold: the text description, obtained by crawling the affiliated metadata of the image, often suffers from the semantic mismatch and the mutual compatibility. To address these issues, we introduce PyramidCLIP, which constructs an input pyramid with different semantic levels for each modality, and aligns visual elements and linguistic elements in the form of hierarchy via peer-level semantics alignment and cross-level relation alignment. Furthermore, we soften the loss of negative samples (unpaired samples) so as to weaken the strict constraint during the pre-training stage, thus mitigating the risk of forcing the model to distinguish compatible negative pairs. Experiments on five downstream tasks demonstrate the effectiveness of the proposed PyramidCLIP. In particular, with the same amount of 15 million pre-training image-text pairs, PyramidCLIP exceeds CLIP on ImageNet zero-shot classification top-1 accuracy by 10.6%/13.2%/10.0% with ResNet50/ViT-B32/ViT-B16 based image encoder respectively. When scaling to larger datasets, PyramidCLIP achieves the state-of-the-art results on several downstream tasks. In particular, the results of PyramidCLIP-ResNet50 trained on 143M image-text pairs surpass that of CLIP using 400M data on ImageNet zero-shot classification task, significantly improving the data efficiency of CLIP.

cs.CV↗

arXiv2022

Evolution as a Service: A Privacy-Preserving Genetic Algorithm for Combinatorial Optimization

Evolutionary algorithms (EAs), such as the genetic algorithm (GA), offer an elegant way to handle combinatorial optimization problems (COPs). However, limited by expertise and resources, most users do not have enough capability to implement EAs to solve COPs. An intuitive and promising solution is to outsource evolutionary operations to a cloud server, whilst it suffers from privacy concerns. To this end, this paper proposes a novel computing paradigm, evolution as a service (EaaS), where a cloud server renders evolutionary computation services for users without sacrificing users' privacy. Inspired by the idea of EaaS, this paper designs PEGA, a novel privacy-preserving GA for COPs. Specifically, PEGA enables users outsourcing COPs to the cloud server holding a competitive GA and approximating the optimal solution in a privacy-preserving manner. PEGA features the following characteristics. First, any user without expertise and enough resources can solve her COPs. Second, PEGA does not leak contents of optimization problems, i.e., users' privacy. Third, PEGA has the same capability as the conventional GA to approximate the optimal solution. We implements PEGA falling in a twin-server architecture and evaluates it in the traveling salesman problem (TSP, a widely known COP). Particularly, we utilize encryption cryptography to protect users' privacy and carefully design a suit of secure computing protocols to support evolutionary operators of GA on encrypted data. Privacy analysis demonstrates that PEGA does not disclose the contents of the COP to the cloud server. Experimental evaluation results on four TSP datasets show that PEGA is as effective as the conventional GA in approximating the optimal solution.

cs.NE↗

arXiv2022

Schwarzschild quasi-normal modes of non-minimally coupled vector fields

We study perturbations of massive and massless vector fields on a Schwarzschild black-hole background, including a non-minimal coupling between the vector field and the curvature. The coupling is given by the Horndeski vector-tensor operator, which we show to be unique, also when the field is massive, provided that the vector has a vanishing background value. We determine the quasi-normal mode spectrum of the vector field, focusing on the fundamental mode of monopolar and dipolar perturbations of both even and odd parity, as a function of the mass of the field and the coupling constant controlling the non-minimal interaction. In the massless case, we also provide results for the first two overtones, showing in particular that the isospectrality between even and odd modes is broken by the non-minimal gravitational coupling. We also consider solutions to the mode equations corresponding to quasi-bound states and static configurations. Our results for quasi-bound states provide strong evidence for the stability of the spectrum, indicating the impossibility of a vectorization mechanism within our set-up. For static solutions, we analytically and numerically derive results for the electromagnetic susceptibilities (the spin-1 analogs of the tidal Love numbers), which we show to be non-zero in the presence of the non-minimal coupling.

gr-qc↗

arXiv2022

Massive Star-Forming Galaxies Have Converted Most of Their Halo Gas into Stars

In the local Universe, the efficiency for converting baryonic gas into stars is very low. In dark matter halos where galaxies form and evolve, the average efficiency varies with galaxy stellar mass and has a maximum of about twenty percent for Milky-Way-like galaxies. The low efficiency at higher mass is believed to be produced by some quenching processes, such as the feedback from active galactic nuclei. We perform an analysis of weak lensing and satellite kinematics for SDSS central galaxies. Our results reveal that the efficiency is much higher, more than sixty percent, for a large population of massive star-forming galaxies around $10^{11}M_{\odot}$. This suggests that these galaxies acquired most of the gas in their halos and converted it into stars without being affected significantly by quenching processes. This population of galaxies is not reproduced in current galaxy formation models, indicating that our understanding of galaxy formation is incomplete. The implications of our results on circumgalactic media, star formation quenching and disc galaxy rotation curves are discussed. We also examine systematic uncertainties in halo-mass and stellar-mass measurements that might influence our results.

astro-ph.GA↗

arXiv2022

Photoelastic Stress Response of Complex 3D-Printed Particle Shapes

While stress visualization within 3-dimensional particles would greatly advance our understanding of the behaviors of complex particles, traditional photoelastic methods suffer from a lack of available technology for producing suitable complex particles. Recently, 3D-printing has created new possibilities for enhancing the scope of stress analysis within physically representative granules. Here, we investigate and evaluate opportunities offered by 3D-printing a single particle with a complex external shape with photoelastic properties. We report the results of X-ray computed tomography and 3D-printing, combined with traditional photoelastic analysis, to visualize strain for particles ranging from simple 2D discs to complex 3D printed coffee beans, including with internal voids. We find that the relative orientation of the print layers and the loading force affects the optical response of the discs, but without a significant difference in their mechanical properties. Furthermore, we present semi-quantitative measurements of stresses within 3D-printed complex particles. The paper outlines the potential limitations and areas of future interest for stress visualization of 3-dimensional particles.

cond-mat.soft↗

arXiv2022

Error Rate Analysis for Grant-free Massive Random Access with Short-Packet Transmission

Grant-free massive random access (RA) is a promising protocol to support the massive machine-type communications (mMTC) scenario in 5G and beyond networks. In this paper, we focus on the error rate analysis in grant-free massive RA, which is critical for practical deployment but has not been well studied. We consider a two-phase frame structure, with a pilot transmission phase for activity detection and channel estimation, followed by a data transmission phase with coded data symbols. Considering the characteristics of short-packet transmission, we analyze the block error rate (BLER) in the finite blocklength regime to characterize the data transmission performance. The analysis involves characterizing the activity detection and channel estimation errors as well as applying the random matrix theory (RMT) to analyze the distribution of the post-processing signal-to-noise ratio (SNR). As a case study, the derived BLER expression is further simplified to optimize the pilot length. Simulation results verify our analysis and demonstrate its effectiveness in pilot length optimization.

cs.IT↗

arXiv2022

First measurement of the characteristic depletion radius of dark matter haloes from weak lensing

We use weak lensing observations to make the first measurement of the characteristic depletion radius, one of the three radii that characterize the region where matter is being depleted by growing haloes. The lenses are taken from the halo catalog produced by the extended halo-based group/cluster finder applied to DESI Legacy Imaging Surveys DR9, while the sources are extracted from the DECaLS DR8 imaging data with the Fourier_Quad pipeline. We study halo masses $12 < \log ( M_{\rm grp} ~[{\rm M_{\odot}}/h] ) \leq 15.3$ within redshifts $0.2 \leq z \leq 0.3$. The virial and splashback radii are also measured and used to test the original findings on the depletion region. When binning haloes by mass, we find consistency between most of our measurements and predictions from the CosmicGrowth simulation, with exceptions to the lowest mass bins. The characteristic depletion radius is found to be roughly $2.5$ times the virial radius and $1.7 - 3$ times the splashback radius, in line with an approximately universal outer density profile, and the average enclosed density within the characteristic depletion radius is found to be roughly $29$ times the mean matter density of the Universe in our sample. When binning haloes by both mass and a proxy for halo concentration, we do not detect a significant variation of the depletion radius with concentration, on which the simulation prediction is also sensitive to the choice of concentration proxy. We also confirm that the measured splashback radius varies with concentration differently from simulation predictions.

astro-ph.CO↗