arXiv subjects

Cheng Li

Publications and source records attributed to Cheng Li.

At least 307 records · Page 17Linked to original sources

arXiv2022

SelfCoLearn: Self-supervised collaborative learning for accelerating dynamic MR imaging

Lately, deep learning has been extensively investigated for accelerating dynamic magnetic resonance (MR) imaging, with encouraging progresses achieved. However, without fully sampled reference data for training, current approaches may have limited abilities in recovering fine details or structures. To address this challenge, this paper proposes a self-supervised collaborative learning framework (SelfCoLearn) for accurate dynamic MR image reconstruction from undersampled k-space data. The proposed framework is equipped with three important components, namely, dual-network collaborative learning, reunderampling data augmentation and a specially designed co-training loss. The framework is flexible to be integrated with both data-driven networks and model-based iterative un-rolled networks. Our method has been evaluated on in-vivo dataset and compared it to four state-of-the-art methods. Results show that our method possesses strong capabilities in capturing essential and inherent representations for direct reconstructions from the undersampled k-space data and thus enables high-quality and fast dynamic MR imaging.

eess.IV↗

arXiv2022

Jupiter's Temperature Structure: A Reassessment of the Voyager Radio Occultation Measurements

The thermal structure of planetary atmospheres is an essential input for predicting and retrieving the distribution of gases and aerosols, as well as the bulk chemical abundances. In the case of Jupiter, the temperature at a reference level - generally taken at 1 bar - serves as the anchor in models used to derive the planet's interior structure and composition. Most models assume the temperature measured by the Galileo probe (Seiff et al. 1998). However, those data correspond to a single location, an unusually clear, dry region, affected by local atmospheric dynamics. On the other hand, the Voyager radio occultation observations cover a wider range of latitudes, longitudes, and times (Lindal et al. 1981). The Voyager retrievals were based on atmospheric composition and radio refractivity data that require updating and were never properly tabulated: the few existing tabulations are incomplete and ambiguous. Here, we present a systematic electronic digitization of all available temperature profiles from Voyager, followed by their reanalysis, employing currently accepted values of the abundances and radio refractivities of atmospheric species. We find the corrected temperature at the 1 bar level to be up to 4 K greater than previously published values, i.e., 170.3{\pm}3.8 K at 12°S (Voyager 1 ingress) and 167.3{\pm}3.8 K at 0°N (Voyager 1 egress). This is to be compared with the Galileo probe value of 166.1{\pm}0.8 K at the edge of an unusual feature at 6.57°N. Altogether, this suggests that Jupiter's tropospheric temperatures may vary spatially by up to 7 K between 7°N and 12°S.

astro-ph.EP↗

arXiv2022

Rethinking Query-Key Pairwise Interactions in Vision Transformers

Vision Transformers have achieved state-of-the-art performance in many visual tasks. Due to the quadratic computational and memory complexities of self-attention, recent works either apply attention only to low-resolution inputs or restrict the receptive field to a small local region. To overcome these limitations, we propose key-only attention, which excludes query-key pairwise interactions and uses a compute-efficient saliency-gate to obtain attention weights, modeling local-global interactions in all stages. Key-only attention has linear computational and memory complexities w.r.t input size. We use alternate layout to hybridize convolution and attention layers instead of grafting which is suggested by previous works, so that all stages can benefit from both spatial attentions and convolutions. We leverage these improvements to develop a new self-attention model family, LinGlos, which reach state-of-the-art accuracies on the parameter-limited setting of ImageNet classification benchmark, and outperform baselines significantly in downstream tasks, e.g., COCO object detection and ADE20K semantic segmentation.

cs.CV↗

arXiv2022

DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale

The past several years have witnessed the success of transformer-based models, and their scale and application scenarios continue to grow aggressively. The current landscape of transformer models is increasingly diverse: the model size varies drastically with the largest being of hundred-billion parameters; the model characteristics differ due to the sparsity introduced by the Mixture-of-Experts; the target application scenarios can be latency-critical or throughput-oriented; the deployment hardware could be single- or multi-GPU systems with different types of memory and storage, etc. With such increasing diversity and the fast-evolving pace of transformer models, designing a highly performant and efficient inference system is extremely challenging. In this paper, we present DeepSpeed Inference, a comprehensive system solution for transformer model inference to address the above-mentioned challenges. DeepSpeed Inference consists of (1) a multi-GPU inference solution to minimize latency while maximizing the throughput of both dense and sparse transformer models when they fit in aggregate GPU memory, and (2) a heterogeneous inference solution that leverages CPU and NVMe memory in addition to the GPU memory and compute to enable high inference throughput with large models which do not fit in aggregate GPU memory. DeepSpeed Inference reduces latency by up to 7.3X over the state-of-the-art for latency-oriented scenarios and increases throughput by over 1.5x for throughput-oriented scenarios. Moreover, it enables trillion parameter scale inference under real-time latency constraints by leveraging hundreds of GPUs, an unprecedented scale for inference. It can inference 25x larger models than with GPU-only solutions, while delivering a high throughput of 84 TFLOPS (over $50\%$ of A6000 peak).

cs.LG↗

arXiv2022

Idealized 2D Cloud-Resolving Simulations for Tidally Locked Habitable Planets

Cloud is critical for planetary climate and habitability, but it is also one of the most challenging parts of studying planets in and beyond the solar system. Here we use a cloud-resolving model (CRM) with high resolution (2 km) in a two-dimensional (2D) configuration to simulate the clouds and circulation on tidally locked aqua-planets. We find that the substellar area is covered by deep convective clouds, the nightside is dominated by low-level clouds, and these two are linked by a global-scale Walker circulation. We further find that a uniform surface warming causes the substellar cloud width to decrease, but a reduction in day-night surface temperature contrast or an increase in longwave radiative cooling rate causes the substellar cloud width to increase. These relationships can be roughly interpreted based on simple thermodynamic theories. Comparing the results between CRM and global 3D general circulation model (GCM), we find that they show qualitatively consistent results, including the Walker circulation, the substellar clouds, and the responses of the substellar ascending area and strength to changes in surface temperature or in its zonal contrast. But, large quantitative differences exist, such as the magnitude of cloud water path, the cloud width, and their responses to external forcings. These results increase our confidence in using GCMs for modeling exoplanetary climates, although large quantitative uncertainties should always exist. Future work is required to use 3D CRM(s) with realistic radiative transfer and with the Coriolis force to examine the clouds and climate of tidally locked planets.

astro-ph.EP↗

arXiv2022

Efficient and ultra-stable perovskite light-emitting diodes

Perovskite light-emitting diodes (PeLEDs) have emerged as a strong contender for next-generation display and information technologies. However, similar to perovskite solar cells, the poor operational stability remains the main obstacle toward commercial applications. Here we demonstrate ultra-stable and efficient PeLEDs with extraordinary operational lifetimes (T50) of 1.0x10^4 h, 2.8x10^4 h, 5.4x10^5 h, and 1.9x10^6 h at initial radiance (or current densities) of 3.7 W/sr/m2 (~5 mA/cm2), 2.1 W/sr/m2 (~3.2 mA/cm2), 0.42 W/sr/m2 (~1.1 mA/cm2), and 0.21 W/sr/m2 (~0.7 mA/cm2) respectively, and external quantum efficiencies of up to 22.8%. Key to this breakthrough is the introduction of a dipolar molecular stabilizer, which serves two critical roles simultaneously. First, it prevents the detrimental transformation and decomposition of the alpha-phase FAPbI3 perovskite, by inhibiting the formation of lead and iodide intermediates. Secondly, hysteresis-free device operation and microscopic luminescence imaging experiments reveal substantially suppressed ion migration in the emissive perovskite. The record-long PeLED lifespans are encouraging, as they now satisfy the stability requirement for commercial organic LEDs (OLEDs). These results remove the critical concern that halide perovskite devices may be intrinsically unstable, paving the path toward industrial applications.

cond-mat.mtrl-sci↗

arXiv2022

Seasonal variations of chemical species and haze in Titan's upper atmosphere

Seasonal variation is significant in Titan's atmosphere due to the large change of solar insolation resulting from Titan's 26.7° axial tilt relative to the plane of Saturn's orbit. Here we present an investigation of hydrocarbon and nitrile species in Titan's upper atmosphere at 400-1200 km, which includes the mesosphere and the lower thermosphere, over more than one fourth of Titan's year (2006-2014, LS=318°-60°), using eighteen stellar occultation observations obtained by Cassini/UVIS. Vertical profiles of eight chemical species (CH4, C2H2, C2H4, C2H6, C4H2, C6H6, HCN, HC3N) and haze particles are retrieved from these observations using an instrument forward model, which considers the technical issue of pointing motion. The Markov-chain Monte Carlo (MCMC) algorithm is used to obtain the posterior probability distributions of parameters in the retrieval, which inherently tests the extent to which species profiles can be constrained. The results show that no change of the species profiles is noticeable before the equinox, while the decrease of atmospheric temperature and significant upwelling in the summer hemisphere are found five terrestrial years afterwards. Altitude of the detached haze layer decreases towards the vernal equinox then it disappears, and no reappearance is identified within the time range of our data, which is consistent with observations from Cassini/ISS. This study provides observational constraints on the seasonal change of Titan's upper atmosphere, and suggests further investigations of the atmospheric chemistry and dynamics therein.

astro-ph.EP↗

arXiv2022

Energetic stability and spatial inhomogeneity in the local electronic structure of relaxed twisted trilayer graphene

We study the energetic stability and the local electronic structure of the general twisted trilayer graphene (TTG) with the top and bottom layers rotated with respect to the middle layer respectively by $θ$ and $θ'$. Approximate supercells of the moiré-of-moiré superlattices with $θ$ and $θ^{\prime}$ within $1^{\circ}\sim 2^{\circ}$ are established to describe the structural and electronic properties of relaxed TTG with the periodic boundary condition. Full relaxation demonstrates that the commensurate TTG with $θ=θ^{\prime}$ has the local minimum total energy ($E_{tol}$) at a fixed $θ$, while $E_{tol}$ first reaches a local maximum and begins to drop with decreasing $θ^{\prime}$ for $θ^{\prime} < θ$. Some regions exhibit enhanced in-plane relaxation in the top and bottom layers but suppressed relaxation in the middle layer and form a hexagonal network with the moiré-of-moiré length scale. The stacking configurations with the atoms in the three layers vertically aligned at the origin of the relaxed TTG supercells at $θ$ around $1.6^{\circ}$ and $θ^{\prime}$ around $1.4^{\circ}$ have a high density of states (DOS) near the Fermi level ($E_F$), which can reach that of the mirror symmetric TTG with equal twist angles of about $1.7^{\circ}$. In contrast, some other stackings can have rather low DOS around $E_F$. The significant stacking dependence of DOS for some TTG supercells demonstrates that the local electronic structure of TTG can exhibit strong spatial inhomogeneity when the twist angles are slightly away from those of the small supercells with large variations of DOS among different stackings. Moreover, the structural relaxation of TTG plays a crucial role in the high DOS and its strong stacking dependence.

cond-mat.mes-hall↗

arXiv2022

Rethinking the optimization process for self-supervised model-driven MRI reconstruction

Recovering high-quality images from undersampled measurements is critical for accelerated MRI reconstruction. Recently, various supervised deep learning-based MRI reconstruction methods have been developed. Despite the achieved promising performances, these methods require fully sampled reference data, the acquisition of which is resource-intensive and time-consuming. Self-supervised learning has emerged as a promising solution to alleviate the reliance on fully sampled datasets. However, existing self-supervised methods suffer from reconstruction errors due to the insufficient constraint enforced on the non-sampled data points and the error accumulation happened alongside the iterative image reconstruction process for model-driven deep learning reconstrutions. To address these challenges, we propose K2Calibrate, a K-space adaptation strategy for self-supervised model-driven MR reconstruction optimization. By iteratively calibrating the learned measurements, K2Calibrate can reduce the network's reconstruction deterioration caused by statistically dependent noise. Extensive experiments have been conducted on the open-source dataset FastMRI, and K2Calibrate achieves better results than five state-of-the-art methods. The proposed K2Calibrate is plug-and-play and can be easily integrated with different model-driven deep learning reconstruction methods.

eess.IV↗

arXiv2022

Correlation between nuclear temperatures and symmetry energy in sub-saturation

A study of the correlation between nuclear temperatures and symmetry energy is presented for heavy-ion collisions at intermediate energies via the isospin-dependent quantum molecular-dynamics model. It is found that different symmetry energy parameters will change the density and kinetic energy distribution of the hot nuclei. More importantly, nuclear temperatures that are based on kinetic energy properties can be used to study symmetry energy information.

nucl-th↗

arXiv2022

Weighted Sum-Rate Maximization for Rate-Splitting Multiple Access Based Secure Communication

As investigations on physical layer security evolve from point-to-point systems to multi-user scenarios, multi-user interference (MUI) is introduced and becomes an unavoidable issue. Different from treating MUI totally as noise in conventional secure communications, in this paper, we propose a rate-splitting multiple access (RSMA)-based secure beamforming design, where user messages are split and encoded into common and private streams. Each user not only decodes the common stream and the intended private stream, but also tries to eavesdrop the private streams of other users. We formulate a weighted sum-rate (WSR) maximization problem subject to the secrecy rate requirements of all users. To tackle the non-convexity of the formulated problem, a successive convex approximation (SCA)-based approach is adopted to convert the original non-convex and intractable problem into a low-complexity suboptimal iterative algorithm. Numerical results demonstrate that the proposed secure beamforming scheme outperforms the conventional multi-user linear precoding (MULP) technique in terms of the WSR performance while ensuring user secrecy rate requirements.

cs.IT↗

arXiv2022

Distributed Bayesian Varying Coefficient Modeling Using a Gaussian Process Prior

Varying coefficient models (VCMs) are widely used for estimating nonlinear regression functions for functional data. Their Bayesian variants using Gaussian process priors on the functional coefficients, however, have received limited attention in massive data applications, mainly due to the prohibitively slow posterior computations using Markov chain Monte Carlo (MCMC) algorithms. We address this problem using a divide-and-conquer Bayesian approach. We first create a large number of data subsamples with much smaller sizes. Then, we formulate the VCM as a linear mixed-effects model and develop a data augmentation algorithm for obtaining MCMC draws on all the subsets in parallel. Finally, we aggregate the MCMC-based estimates of subset posteriors into a single Aggregated Monte Carlo (AMC) posterior, which is used as a computationally efficient alternative to the true posterior distribution. Theoretically, we derive minimax optimal posterior convergence rates for the AMC posteriors of both the varying coefficients and the mean regression function. We provide quantification on the orders of subset sample sizes and the number of subsets. The empirical results show that the combination schemes that satisfy our theoretical assumptions, including the AMC posterior, have better estimation performance than their main competitors across diverse simulations and in a real data analysis.

stat.ME↗

arXiv2022

Expert Knowledge-guided Geometric Representation Learning for Magnetic Resonance Imaging-based Glioma Grading

Radiomics and deep learning have shown high popularity in automatic glioma grading. Radiomics can extract hand-crafted features that quantitatively describe the expert knowledge of glioma grades, and deep learning is powerful in extracting a large number of high-throughput features that facilitate the final classification. However, the performance of existing methods can still be improved as their complementary strengths have not been sufficiently investigated and integrated. Furthermore, lesion maps are usually needed for the final prediction at the testing phase, which is very troublesome. In this paper, we propose an expert knowledge-guided geometric representation learning (ENROL) framework . Geometric manifolds of hand-crafted features and learned features are constructed to mine the implicit relationship between deep learning and radiomics, and therefore to dig mutual consent and essential representation for the glioma grades. With a specially designed manifold discrepancy measurement, the grading model can exploit the input image data and expert knowledge more effectively in the training phase and get rid of the requirement of lesion segmentation maps at the testing phase. The proposed framework is flexible regarding deep learning architectures to be utilized. Three different architectures have been evaluated and five models have been compared, which show that our framework can always generate promising results.

eess.IV↗

arXiv2021

High dynamic range base design and characterization of an 8-inch photomultiplier tube CR365-02-2 for the LHAASO-MD experiment

Ultra-high-energy ($>$ 100 TeV) gamma-ray detection benefits from the use of muon detectors (MDs) thanks to their capability of suppressing the cosmic-ray background. More than 1100 8-inch photomultiplier tubes (PMTs), CR365-02-2 from Beijing Hamamatsu Photon Techniques INC. (BHP), will be deployed for the LHAASO-MD experiment. In this paper, the design of the photomultiplier base for CR365-02-2 is presented. Signals are extracted from two outputs: the anode and the 7th dynode. This design ensures a high dynamic range, from single photoelectron (SPE, peak-to-valley ratio $>$ 2), up to bunches of $7\times10^5$ photoelectrons, with an equivalent anode peak current of up to 1600 mA. The anode-to-dynode amplitude ratio (A/D) is below 160 in order to ensure enough overlap between the two outputs. Test results of 4 PMTs, including the SPE response, A/D, linearity, dark count rate, and afterpulse rate, are reported.

physics.ins-det↗

arXiv2021

Radiomic biomarker extracted from PI-RADS 3 patients support more eìcient and robust prostate cancer diagnosis: a multi-center study

Prostate Imaging Reporting and Data System (PI-RADS) based on multi-parametric MRI classiêes patients into 5 categories (PI-RADS 1-5) for routine clinical diagnosis guidance. However, there is no consensus on whether PI-RADS 3 patients should go through biopsies. Mining features from these hard samples (HS) is meaningful for physicians to achieve accurate diagnoses. Currently, the mining of HS biomarkers is insuìcient, and the eéectiveness and robustness of HS biomarkers for prostate cancer diagnosis have not been explored. In this study, biomarkers from diéerent data distributions are constructed. Results show that HS biomarkers can achieve better performances in diéerent data distributions.

eess.IV↗

arXiv2021

Peta-electron volt gamma-ray emission from the Crab Nebula

The Crab pulsar and the surrounding nebula powered by the pulsar's rotational energy through the formation and termination of a relativistic electron-positron wind is a bright source of gamma-rays carrying crucial information about this complex conglomerate. We report the detection of $γ$-rays with a spectrum showing gradual steepening over three energy decades, from $5\times 10^{-4}$ to $1.1$ petaelectronvolt (PeV). The ultra-high-energy photons exhibit the presence of a PeV electron accelerator (a pevatron) with an acceleration rate exceeding 15% of the absolute theoretical limit. Assuming that unpulsed $γ$-rays are produced at the termination of the pulsar's wind, we constrain the pevatron's size, between $0.025$ and $0.1$ pc, and the magnetic field $\approx 110 μ$G. The production rate of PeV electrons, $2.5 \times 10^{36}$ erg $\rm s^{-1}$, constitutes 0.5% of the pulsar's spin-down luminosity, although we do not exclude a non-negligible contribution of PeV protons to the production of the highest energy $γ$-rays.

astro-ph.HE↗

arXiv2021

An IFU View of the Active Galactic Nuclei in MaNGA Galaxy Pairs

The role of active galactic nuclei (AGNs) during galaxy interactions and how they influence the star formation in the system are still under debate. We use a sample of 1156 galaxies in galaxy pairs or mergers (hereafter `pairs') from the MaNGA survey. This pair sample is selected by the velocity offset, projected separation, and morphology, and is further classified into four cases along the merger sequence based on morphological signatures. We then identify a total of 61 (5.5%) AGNs in pairs based on the emission-line diagnostics. No evolution of the AGN fraction is found, either along the merger sequence or compared to isolated galaxies (5.0%). We observe a higher fraction of passive galaxies in galaxy pairs, especially in the pre-merging cases, and associate the higher fraction to their environmental dependence. The isolated AGN and AGN in pairs show similar distributions in their global stellar mass, star formation rate (SFR), and central [OIII] surface brightness. AGNs in pairs show radial profiles of increasing specific SFR and declining Dn4000 from center to outskirts, and no significant difference from the isolated AGNs. This is clearly different from star-forming galaxies (SFGs) in our pair sample, which show enhanced central star formation, as reported before. AGNs in pairs have lower Balmer decrements at outer regions, possibly indicating less dust attenuation. Our findings suggest that AGNs likely follow an inside-out quenching and the merger impact on the star formation in AGNs is less prominent than in SFGs.

astro-ph.GA↗

arXiv2021

Evidence for multiple Ferrel-like cells on Jupiter

Jupiter's atmosphere is dominated by multiple jet streams which are strongly tied to its 3D atmospheric circulation. Lacking a rigid bottom boundary, several models exist for how the meridional circulation extends into the planetary interior. Here we show, collecting evidence from multiple instruments of the Juno mission, the existence of mid-latitudinal meridional circulation cells which are driven by turbulence, similar to the Ferrel cells on Earth. Different than Earth, which contains only one such cell in each hemisphere, the larger, faster rotating Jupiter can incorporate multiple cells. The cells form regions of upwelling and downwelling, which we show are clearly evident in Juno's microwave data between latitude 60S and 60N. The existence of these cells is confirmed by reproducing the ammonia observations using a simplistic model. This study solves a long-standing puzzle regarding the nature of Jupiter's sub-cloud dynamics and provides evidence for 8 cells in each Jovian hemisphere.

astro-ph.EP↗