Searcharxiv⌕ Search

arXiv subjects

Chao Yang

Publications and source records attributed to Chao Yang.

At least 289 records · Page 16Linked to original sources

End-to-end Adaptive Distributed Training on PaddlePaddle

Distributed training has become a pervasive and effective approach for training a large neural network (NN) model with processing massive data. However, it is very challenging to satisfy requirements from various NN models, diverse computing resources, and their dynamic changes during a training job. In this study, we design our distributed training framework in a systematic end-to-end view to provide the built-in adaptive ability for different scenarios, especially for industrial applications and production environments, by fully considering resource allocation, model partition, task placement, and distributed execution. Based on the unified distributed graph and the unified cluster object, our adaptive framework is equipped with a global cost model and a global planner, which can enable arbitrary parallelism, resource-aware placement, multi-mode execution, fault-tolerant, and elastic distributed training. The experiments demonstrate that our framework can satisfy various requirements from the diversity of applications and the heterogeneity of resources with highly competitive performance. The ERNIE language model with 260 billion parameters is efficiently trained on thousands of AI processors with 91.7% weak scalability. The throughput of the model from the recommender system by employing the heterogeneous pipeline asynchronous execution can be increased up to 2.1 times and 3.3 times that of the GPU-only and CPU-only training respectively. Moreover, the fault-tolerant and elastic distributed training have been successfully applied to the online industrial applications, which give a reduction of 34.49% in the number of failed long-term training jobs and an increase of 33.91% for the global scheduling efficiency in the production environment.

cs.DC↗

A rank-adaptive higher-order orthogonal iteration algorithm for truncated Tucker decomposition

We propose a novel rank-adaptive higher-order orthogonal iteration (HOOI) algorithm to compute the truncated Tucker decomposition of higher-order tensors with a given error tolerance, and prove that the method is locally optimal and monotonically convergent. A series of numerical experiments related to both synthetic and real-world tensors are carried out to show that the proposed rank-adaptive HOOI algorithm is advantageous in terms of both accuracy and efficiency. Some further analysis on the HOOI algorithm and the classical alternating least squares method are presented to further understand why rank adaptivity can be introduced into the HOOI algorithm and how it works.

math.NA↗

Investigating the co-evolution of massive black holes in dual active galactic nuclei and their host galaxies via galaxy merger simulations

Major galaxy mergers can trigger nuclear activities and are responsible for high-luminosity quasi-stellar objects /active galactic nuclei (QSOs/AGNs). In certain circumstances, such mergers may cause dual active galactic nuclei (dAGN) phenomenon. This study investigates dAGN triggering and evolution of massive black holes (MBHs) during the merging processes using hydrodynamic code GADGET-2 to simulate several gas-rich major mergers at redshift $z=2$ and $3$, respectively. Results reveal that gas-rich major mergers can trigger significant nuclear activities after the second and third pericentric passages and the formation of dAGN with significant time duration ($\sim 10 - 390$ Myr). During the merging processes, galactic bulge evolves with time because of the rapid star formation in each (or both) galactic centers and initial mixing of stars in galactic disks due to violent relaxation. MBHs grow substantially due to accretion and finally merge into a bigger black hole. The growth of galactic bulges and corresponding increases of its velocity dispersions predate the growth of MBHs in the dAGN stages. The MBHs in these stages deviate below the relation between MBH mass and bulge mass (or velocity dispersion), and they revert to the relation after the final mergers due to the significant accretion that occurs mostly at a separation less than a few kpc. Then, the two MBHs merge with each other.

astro-ph.GA↗

Hydrodynamical Simulations of the Triggering of Nuclear Activities by Minor Mergers of Galaxies

Major mergers of galaxies are considered to be an efficient way to trigger Active Galactic Nuclei and are thought to be responsible for the phenomenon of quasars. This has however recently been challenged by observations of a large number of low luminosity Active Galactic Nuclei at low redshift ($z\lesssim1$) without obvious major merger signatures. Minor mergers are frequently proposed to explain the existence of these Active Galactic Nuclei. In this paper, we perform nine high resolution hydrodynamical simulations of minor galaxy mergers and investigate whether nuclear activities can be efficiently triggered by minor mergers, by setting various properties for the progenitor galaxies of those mergers. We find that minor galaxy merger scan activate the massive black hole in the primary galaxy with an Eddington ratio of $f_{\rm Edd}>0.01$ and $>0.05$ (or a bolometric luminosity $>10^{43}$ and $>10^{44}\mathrm{erg\, s^{-1}}$) with a duration of $2.71$ and $0.49$ Gyr (or $2.69$ and $0.19$ Gyr), respectively. The nuclear activity of primary galaxy strongly depends on the nucleus separation, the nucleus is more active as the two nuclei approach to each other. Dual Active Galactic Nuclei systems can still possibly form by minor mergers of galaxies, the time period for dual Active Galactic Nuclei is only $\sim 0.011$ Gyr and $\sim 0.017$ Gyr with Eddington ratio of $f_{\rm Edd}>0.05$ and bolometric luminosity $>10^{44}\mathrm{erg\, s^{-1}}$. This time period is typically shorter than that of dual Active Galactic Nuclei induced by galaxy major mergers.

astro-ph.GA↗

Accelerating quantum many-body configuration interaction with directives

Many-Fermion Dynamics-nuclear, or MFDn, is a configuration interaction (CI) code for nuclear structure calculations. It is a platform-independent Fortran 90 code using a hybrid MPI+X programming model. For CPU platforms the application has a robust and optimized OpenMP implementation for shared memory parallelism. As part of the NESAP application readiness program for NERSC's latest Perlmutter system, MFDn has been updated to take advantage of accelerators. The current mainline GPU port is based on OpenACC. In this work we describe some of the key challenges of creating an efficient GPU implementation. Additionally, we compare the support of OpenMP and OpenACC on AMD and NVIDIA GPUs.

cs.DC↗

A review of computational tools for generating metagenome-assembled genomes from metagenomic sequencing data

Microbes are essentially yet convolutedly linked with human lives on the earth. They critically interfere in different physiological processes and thus influence overall health status. Studying microbial species is used to be constrained to those that can be cultured in the lab. But it excluded a huge portion of the microbiome that could not survive on lab conditions. In the past few years, the culture-independent metagenomic sequencing enabled us to explore the complex microbial community coexisting within and on us. Metagenomics has equipped us with new avenues of investigating the microbiome, from studying a single species to a complex community in a dynamic ecosystem. Thus, identifying the involved microbes and their genomes becomes one of the core tasks in metagenomic sequencing. Metagenome-assembled genomes are groups of contigs with similar sequence characteristics from de novo assembly and could represent the microbial genomes from metagenomic sequencing. In this paper, we reviewed a spectrum of tools for producing and annotating metagenome-assembled genomes from metagenomic sequencing data and discussed their technical and biological perspectives.

q-bio.GN↗

Accelerating an Iterative Eigensolver for Nuclear Structure Configuration Interaction Calculations on GPUs using OpenACC

To accelerate the solution of large eigenvalue problems arising from many-body calculations in nuclear physics on distributed-memory parallel systems equipped with general-purpose Graphic Processing Units (GPUs), we modified a previously developed hybrid MPI/OpenMP implementation of an eigensolver written in FORTRAN 90 by using an OpenACC directives based programming model. Such an approach requires making minimal changes to the original code and enables a smooth migration of large-scale nuclear structure simulations from a distributed-memory many-core CPU system to a distributed GPU system. However, in order to make the OpenACC based eigensolver run efficiently on GPUs, we need to take into account the architectural differences between a many-core CPU and a GPU device. Consequently, the optimal way to insert OpenACC directives may be different from the original way of inserting OpenMP directives. We point out these differences in the implementation of sparse matrix-matrix multiplications (SpMM), which constitutes the main cost of the eigensolver, as well as other differences in the preconditioning step and dense linear algebra operations. We compare the performance of the OpenACC based implementation executed on multiple GPUs with the performance on distributed-memory many-core CPUs, and demonstrate significant speedup achieved on GPUs compared to the on-node performance of a many-core CPU. We also show that the overall performance improvement of the eigensolver on multiple GPUs is more modest due to the communication overhead among different MPI ranks.

cs.DC↗

Global well-posedness for fractional Sobolev-Galpern type equations

This article is a comparative study on an initial-boundary value problem for a class of semilinear pseudo-parabolic equations with the fractional Caputo derivative, also called the fractional Sobolev-Galpern type equations. The purpose of this work is to reveal the influence of the degree of the source nonlinearity on the well-posedness of the solution. By considering four different types of nonlinearities, we derive the global well-posedness of mild solutions to the problem corresponding to the four cases of the nonlinear source terms. For the advection source function case, we apply a nontrivial limit technique for singular integral and some appropriate choices of weighted Banach space to prove the global existence result. For the gradient nonlinearity as a local Lipschitzian, we use the Cauchy sequence technique to show that the solution either exists globally in time or blows up at finite time. For the polynomial form nonlinearity, by assuming the smallness of the initial data we derive the global well-posed results. And for the case of exponential nonlinearity in two-dimensional space, we derive the global well-posedness by additionally use of Orlicz space.

math.AP↗

Economic Dispatch of an Integrated Microgrid Based on the Dynamic Process of CCGT Plant

Intra-day economic dispatch of an integrated microgrid is a fundamental requirement to integrate distributed generators. The dynamic energy flows in cogeneration units present challenges to the energy management of the microgrid. In this paper, a novel approximate dynamic programming (ADP) approach is proposed to solve this problem based on value function approximation, which is distinct with the consideration of the dynamic process constraints of the combined-cycle gas turbine (CCGT) plant. First, we mathematically formulate the multi-time periods decision problem as a finite-horizon Markov decision process. To deal with the thermodynamic process, an augmented state vector of CCGT is introduced. Second, the proposed VFA-ADP algorithm is employed to derive the near-optimal real-time operation strategies. In addition, to guarantee the monotonicity of piecewise linear function, we apply the SPAR algorithm in the update process. To validate the effectiveness of the proposed method, we conduct experiments with comparisons to some traditional optimization methods. The results indicate that our proposed ADP method achieves better performance on the economic dispatch of the microgrid.

eess.SY↗

A note on 1-2-3 and 1-2 Conjectures for 3-regular graphs

The 1-2-3 Conjecture, posed by Karoński, Łuczak and Thomason, asked whether every connected graph $G$ different from $K_2$ can be 3-edge-weighted so that every two adjacent vertices of $G$ get distinct sums of incident weights. The 1-2 Conjecture states that if vertices also receive colors and the vertex color is added to the sum of its incident edges, then adjacent vertices can be distinguished using only $\{ 1,2\}$. In this paper we confirm 1-2 Conjecture for 3-regular graphs. Meanwhile, we show that every 3-regular graph can achieve a neighbor sum distinguishing edge coloring by using 4 colors, which answers 1-2-3 Conjecture positively.

math.CO↗

Analyzing and predicting non-equilibrium many-body dynamics via dynamic mode decomposition

Simulating the dynamics of a nonequilibrium quantum many-body system by computing the two-time Green's function associated with such a system is computationally challenging. However, we are often interested in the time diagonal of such a Green's function or time dependent physical observables that are functions of one time. In this paper, we discuss the possibility of using dynamic model decomposition (DMD), a data-driven model order reduction technique, to characterize one-time observables associated with the nonequilibrium dynamics using snapshots computed within a small time window. The DMD method allows us to efficiently predict long time dynamics from a limited number of trajectory samples. We demonstrate the effectiveness of DMD on a model two-band system. We show that, in the equilibrium limit, the DMD analysis yields results that are consistent with those produced from a linear response analysis. In the nonequilibrium case, the extrapolated dynamics produced by DMD is more accurate than a special Fourier extrapolation scheme presented in this paper. We point out a potential pitfall of the standard DMD method caused by insufficient spatial/momentum resolution of the discretization scheme. We show how this problem can be overcome by using a variant of the DMD method known as higher order DMD.

cond-mat.stat-mech↗

Adapting Off-the-Shelf Source Segmenter for Target Medical Image Segmentation

Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a labeled source domain to an unlabeled and unseen target domain, which is usually trained on data from both domains. Access to the source domain data at the adaptation stage, however, is often limited, due to data storage or privacy issues. To alleviate this, in this work, we target source free UDA for segmentation, and propose to adapt an ``off-the-shelf" segmentation model pre-trained in the source domain to the target domain, with an adaptive batch-wise normalization statistics adaptation framework. Specifically, the domain-specific low-order batch statistics, i.e., mean and variance, are gradually adapted with an exponential momentum decay scheme, while the consistency of domain shareable high-order batch statistics, i.e., scaling and shifting parameters, is explicitly enforced by our optimization objective. The transferability of each channel is adaptively measured first from which to balance the contribution of each channel. Moreover, the proposed source free UDA framework is orthogonal to unsupervised learning methods, e.g., self-entropy minimization, which can thus be simply added on top of our framework. Extensive experiments on the BraTS 2018 database show that our source free UDA framework outperformed existing source-relaxed UDA methods for the cross-subtype UDA segmentation task and yielded comparable results for the cross-modality UDA segmentation task, compared with a supervised UDA methods with the source data.

cs.CV↗

Improving Co-registration for Sentinel-1 SAR and Sentinel-2 Optical images

Co-registering the Sentinel-1 SAR and Sentinel-2 optical data of European Space Agency (ESA) is of great importance for many remote sensing applications. However, we find that there are evident misregistration shifts between the Sentinel-1 SAR and Sentinel-2 optical images that are directly downloaded from the official website. To address that, this paper presents a fast and effective registration method for the two types of images. In the proposed method, a block-based scheme is first designed to extract evenly distributed interest points. Then the correspondences are detected by using the similarity of structural features between the SAR and optical images, where the three dimension (3D) phase correlation (PC) is used as the similarity measure for accelerating image matching. Finally, the obtained correspondences are employed to measure the misregistration shifts between the images. Moreover, to eliminate the misregistration, we use some representative geometric transformation models such as polynomial models, projective models, and rational function models for the co-registration of the two types of images, and compare and analyze their registration accuracy under different numbers of control points and different terrains. Six pairs of the Sentinel-1 SAR L1 and Sentinel-2 optical L1C images covering three different terrains are tested in our experiments. Experimental results show that the proposed method can achieve precise correspondences between the images, and the 3rd. Order polynomial achieves the most satisfactory registration results. Its registration accuracy of the flat areas is less than 1.0 10m pixels, and that of the hilly areas is about 1.5 10m pixels, and that of the mountainous areas is between 1.7 and 2.3 10m pixels, which significantly improves the co-registration accuracy of the Sentinel-1 SAR and Sentinel-2 optical images.

eess.IV↗

Nonlinear localized excitations in a topological ferromagnetic honeycomb lattice

We theoretically investigate nonlinear localized excitations in a Heisenberg honeycomb ferromagnet with a second neighbour Dzialozinskii-Moriya interaction, which has been proved to possess the topological band structure. Using the time-dependent variational principle, we obtain the equation of motion for the system in the Glauber coherent-state representation. By means of the semidiscrete multiple-scale method, different types of nonlinear localized excitations are gotten.Our results show that both two-dimensional discrete breathers and bulk gap solitons may occur in the Heisenberg honeycomb ferromagnet.

cond-mat.mes-hall↗

Composite Localization for Human Pose Estimation

The existing human pose estimation methods are confronted with inaccurate long-distance regression or high computational cost due to the complex learning objectives. This work proposes a novel deep learning framework for human pose estimation called composite localization to divide the complex learning objective into two simpler ones: a sparse heatmap to find the keypoint's approximate location and two short-distance offsetmaps to obtain its final precise coordinates. To realize the framework, we construct two types of composite localization networks: CLNet-ResNet and CLNet-Hourglass. We evaluate the networks on three benchmark datasets, including the Leeds Sports Pose dataset, the MPII Human Pose dataset, and the COCO keypoints detection dataset. The experimental results show that our CLNet-ResNet50 outperforms SimpleBaseline by 1.14% with about 1/2 GFLOPs. Our CLNet-Hourglass outperforms the original stacked-hourglass by 4.45% on COCO.

cs.CV↗

FDNet: A Deep Learning Approach with Two Parallel Cross Encoding Pathways for Precipitation Nowcasting

With the goal of predicting the future rainfall intensity in a local region over a relatively short period time, precipitation nowcasting has been a long-time scientific challenge with great social and economic impact. The radar echo extrapolation approaches for precipitation nowcasting take radar echo images as input, aiming to generate future radar echo images by learning from the historical images. To effectively handle complex and high non-stationary evolution of radar echoes, we propose to decompose the movement into optical flow field motion and morphologic deformation. Following this idea, we introduce Flow-Deformation Network (FDNet), a neural network that models flow and deformation in two parallel cross pathways. The flow encoder captures the optical flow field motion between consecutive images and the deformation encoder distinguishes the change of shape from the translational motion of radar echoes. We evaluate the proposed network architecture on two real-world radar echo datasets. Our model achieves state-of-the-art prediction results compared with recent approaches. To the best of our knowledge, this is the first network architecture with flow and deformation separation to model the evolution of radar echoes for precipitation nowcasting. We believe that the general idea of this work could not only inspire much more effective approaches but also be applied to other similar spatiotemporal prediction tasks

cs.LG↗

A strange metal in a bosonic system

Fermi liquid theory forms the basis for our understanding of the majority of metals, which is manifested in the description of transport properties that the electrical resistivity goes as temperature squared in the limit of zero temperature. However, the observations of strange metal states in various quantum materials, notably high-temperature superconductors, bring this spectacularly successful theoretical framework into crisis. When electron scattering rate 1/τ hits its limit, kBT/{\hbar} where {\hbar} is the reduced Planck's constant, T represents absolute temperature and kB denotes Boltzmann's constant, Planckian dissipation occurs and lends strange metals a surprising link to black holes, gravity, and quantum information theory. Here, we show the characteristic signature of strange metallicity arising unprecedentedly in a bosonic system. Our nanopatterned YBa2Cu3O7-δ(YBCO) film arrays reveal T-linear resistance as well as B-linear magnetoresistance over an extended temperature and magnetic field range in a quantum critical region in the phase diagram. Moreover, the slope of the T-linear resistance α_cp appears bounded by α_cp {\approx} h/2e^2 [1/T]_c^onset where T_c^onset is the temperature at which Cooper pairs form, intimating a common scale-invariant transport mechanism corresponding to Planckian dissipation.In contrast to fermionic systems where the temperature and magnetic field dependent scattering rates combine in quadrature of {\hbar}/τ {\approx} {\sqrt} (((k_B T)^2+(μ_B B)^2)), both terms linearly combine in the present bosonic system, i.e. {\hbar}/τ {\approx} (k_B T+[γμ]_B B), where γ is a constant. By extending the reach of strange metal phenomenology to a bosonic system, our results suggest that there is a fundamental principle governing their transport which transcends particle statistics.

cond-mat.supr-con↗

Low Rank Approximation in Simulations of Quantum Algorithms

Simulating quantum algorithms on classical computers is challenging when the system size, i.e., the number of qubits used in the quantum algorithm, is moderately large. However, some quantum algorithms and the corresponding quantum circuits can be simulated efficiently on a classical computer if the input quantum state is a low-rank tensor and all intermediate states of the quantum algorithm can be represented or approximated by low-rank tensors. In this paper, we examine the possibility of simulating a few quantum algorithms by using low-rank canonical polyadic (CP) decomposition to represent the input and all intermediate states of these algorithms. Two rank reduction algorithms are used to enable efficient simulation. We show that some of the algorithms preserve the low-rank structure of the input state and can thus be efficiently simulated on a classical computer. However, the rank of the intermediate states in other quantum algorithms can increase rapidly, making efficient simulation more difficult. To some extent, such difficulty reflects the advantage or superiority of a quantum computer over a classical computer. As a result, understanding the low-rank structure of a quantum algorithm allows us to identify algorithms that can benefit significantly from quantum computers.

cs.CE↗