arXiv subjects

Cheng Liu

Publications and source records attributed to Cheng Liu.

At least 163 records · Page 9Linked to original sources

arXiv2022

Winograd Convolution: A Perspective from Fault Tolerance

Winograd convolution is originally proposed to reduce the computing overhead by converting multiplication in neural network (NN) with addition via linear transformation. Other than the computing efficiency, we observe its great potential in improving NN fault tolerance and evaluate its fault tolerance comprehensively for the first time. Then, we explore the use of fault tolerance of winograd convolution for either fault-tolerant or energy-efficient NN processing. According to our experiments, winograd convolution can be utilized to reduce fault-tolerant design overhead by 27.49\% or energy consumption by 7.19\% without any accuracy loss compared to that without being aware of the fault tolerance

cs.LG↗

arXiv2021

A catalogue of 323 cataclysmic variables from LAMOST DR6

In this work, we present a catalog of cataclysmic variables (CVs) identified from the Sixth Data Release (DR6) of the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST). To single out the CV spectra, we introduce a novel machine-learning algorithm called UMAP to screen out a total of 169,509 H$α$-emission spectra, and obtain a classification accuracy of the algorithm of over 99.6$\%$ from the cross-validation set. We then apply the template matching program PyHammer v2.0 to the LAMOST spectra to obtain the optimal spectral type with metallicity, which helps us identify the chromospherically active stars and potential binary stars from the 169,509 spectra. After visually inspecting all the spectra, we identify 323 CV candidates from the LAMOST database, among them 52 objects are new. We further discuss the new CV candidates in subtypes based on their spectral features, including five DN subtype during outbursts, five NL subtype and four magnetic CVs (three AM Her type and one IP type). We also find two CVs that have been previously identified by photometry, and confirm their previous classification by the LAMOST spectra.

astro-ph.SR↗

arXiv2021

HyCA: A Hybrid Computing Architecture for Fault Tolerant Deep Learning

Hardware faults on the regular 2-D computing array of a typical deep learning accelerator (DLA) can lead to dramatic prediction accuracy loss. Prior redundancy design approaches typically have each homogeneous redundant processing element (PE) to mitigate faulty PEs for a limited region of the 2-D computing array rather than the entire computing array to avoid the excessive hardware overhead. However, they fail to recover the computing array when the number of faulty PEs in any region exceeds the number of redundant PEs in the same region. The mismatch problem deteriorates when the fault injection rate rises and the faults are unevenly distributed. To address the problem, we propose a hybrid computing architecture (HyCA) for fault-tolerant DLAs. It has a set of dot-production processing units (DPPUs) to recompute all the operations that are mapped to the faulty PEs despite the faulty PE locations. According to our experiments, HyCA shows significantly higher reliability, scalability, and performance with less chip area penalty when compared to the conventional redundancy approaches. Moreover, by taking advantage of the flexible recomputing, HyCA can also be utilized to scan the entire 2-D computing array and detect the faulty PEs effectively at runtime.

cs.AR↗

arXiv2021

Prospects for the detection of the prompt very-high-energy emission from $\rmγ$-ray bursts with the High Altitude Detection of Astronomical Radiation experiment

The observation of very-high-energy (VHE, $\rm >10~GeV$) $γ$-ray emission from $\rm γ$-ray bursts (GRBs), especially in the prompt phase, will provide critical information for understanding many aspects of their nature including the physical environment, the relativistic bulk motion, the mechanisms of particle acceleration of GRBs and for studying Lorentz invariance violation, etc. For the afterglow phase, the highest energy photons detected to date by the imaging atmospheric Cherenkov telescopes extend to the TeV regime. However, for the prompt phase, years of efforts in searching for the VHE emission has yielded no statistically significant detections. A wide field-of-view (FOV) and large effective area above tens of GeV are essential for detecting the VHE emissions from GRBs in the prompt phase. The High Altitude Detection of Astronomical Radiation (HADAR) experiment has such merits. In this paper, we report the estimates of its expected annual GRB detection rate, which are obtained by combining the performance of the HADAR instrument with the theoretical calculations based on a phenomenological model to generate the pseudo-GRB population. The expected detectable gamma-ray signal from GRBs above the background is then obtained to give the detection rate. In the spectral model, an extra component is assigned to every GRB event in addition to the Band function. The results indicate that if the energy of the cutoff due to internal absorption is higher than 50 GeV, the detection rate for GRBs for the HADAR experiment is approximately two or three GRBs per year, which varies slightly depending upon the characteristics of the extra component.

astro-ph.HE↗

arXiv2021

Taming Process Variations in CNFET for Efficient Last Level Cache Design

Carbon nanotube field-effect transistors (CNFET) emerge as a promising alternative to CMOS transistors for the much higher speed and energy efficiency, which makes the technology particularly suitable for building the energy-hungry last level cache (LLC). However, the process variations (PVs) in CNFET caused by the imperfect fabrication lead to large timing variation and the worst-case timing dramatically limits the LLC operation speed. Particularly, we observe that the CNFET-based cache latency distribution is closely related to the LLC layouts. For the two typical LLC layouts that have the CNT growth direction aligned to the cache way direction and cache set direction respectively, we proposed variation-aware set aligned (VASA) cache and variation-aware way aligned (VAWA) cache in combination with corresponding cache optimizations such as data shuffling and page mapping to enable low-latency cache for frequently used data. According to our experiments, the optimized LLC reduces the average access latency by 32% and 45% compared to the baseline designs on the two different CNFET layouts respectively while it improves the overall performance by 6\% and 9\% and reduces the energy consumption by 4% and 8% respectively. In addition, with both the architecture induced latency variation and PV incurred latency variation considered in a unified model, we extended the VAWA and VASA cache design for the CNFET-based NUCA and the proposed NUCA achieves both significant performance improvement and energy saving compared to the straightforward variation-aware NUCA.

cs.AR↗

arXiv2021

Site-selective $d^{10}$/$d^0$ substitution in a $S = 1/2$ spin ladder Ba$_2$CuTe$_{1-x}$W$_x$O$_6$ ($0 \leq x \leq 0.3$)

Remarkably, doping isovalent $d^{10}$ and $d^0$ cations onto the $B$'' site in $A_2B$'$B$''O$_6$ double perovskites has the power to direct the magnetic interactions between magnetic $B$' cations. This is due to changes in orbital hybridization, which favors different superexchange pathways, and leads to the formation of alternative magnetic structures depending on whether $B$'' is $d^{10}$ or $d^0$. Furthermore, the competition generated by introducing mixtures of $d^{10}$ and $d^0$ cations can drive the material into the realms of exotic quantum magnetism. Here, a W$^{6+}$ $d^0$ dopant was introduced to a $d^{10}$ hexagonal perovskite Ba$_2$CuTeO$_6$, which possesses a spin ladder geometry of Cu$^{2+}$ cations, creating a Ba$_2$CuTe$_{1-x}$W$_x$O$_6$ solid solution ($x$ = 0 - 0.3). Neutron and synchrotron X-ray diffraction show that W$^{6+}$ is almost exclusively substituted for Te$^{6+}$ on the corner-sharing site within the spin ladder, in preference to the face-sharing site between ladders. This means the intra-ladder interactions are selectively tuned by the $d^0$ cations. Bulk magnetic measurements suggest this suppresses magnetic ordering in a similar manner to that observed for the spin-liquid like material Sr$_2$CuTe$_{1-x}$W$_x$O$_6$. This further demonstrates the utility of $d^{10}$ and $d^0$ dopants as a tool for tuning magnetic ground states in a wide range of perovskites and perovskite-derived structures.

cond-mat.str-el↗

arXiv2021

R2F: A Remote Retraining Framework for AIoT Processors with Computing Errors

AIoT processors fabricated with newer technology nodes suffer rising soft errors due to the shrinking transistor sizes and lower power supply. Soft errors on the AIoT processors particularly the deep learning accelerators (DLAs) with massive computing may cause substantial computing errors. These computing errors are difficult to be captured by the conventional training on general purposed processors like CPUs and GPUs in a server. Applying the offline trained neural network models to the edge accelerators with errors directly may lead to considerable prediction accuracy loss. To address the problem, we propose a remote retraining framework (R2F) for remote AIoT processors with computing errors. It takes the remote AIoT processor with soft errors in the training loop such that the on-site computing errors can be learned with the application data on the server and the retrained models can be resilient to the soft errors. Meanwhile, we propose an optimized partial TMR strategy to enhance the retraining. According to our experiments, R2F enables elastic design trade-offs between the model accuracy and the performance penalty. The top-5 model accuracy can be improved by 1.93%-13.73% with 0%-200% performance penalty at high fault error rate. In addition, we notice that the retraining requires massive data transmission and even dominates the training time, and propose a sparse increment compression approach for the data transmission optimization, which reduces the retraining time by 38%-88% on average with negligible accuracy loss over a straightforward remote retraining.

cs.AR↗

arXiv2021

Energy-Efficient Accelerator Design for Deformable Convolution Networks

Deformable convolution networks (DCNs) proposed to address the image recognition with geometric or photometric variations typically involve deformable convolution that convolves on arbitrary locations of input features. The locations change with different inputs and induce considerable dynamic and irregular memory accesses which cannot be handled by classic neural network accelerators (NNAs). Moreover, bilinear interpolation (BLI) operation that is required to obtain deformed features in DCNs also cannot be deployed on existing NNAs directly. Although a general purposed processor (GPP) seated along with classic NNAs can process the deformable convolution, the processing on GPP can be extremely slow due to the lack of parallel computing capability. To address the problem, we develop a DCN accelerator on existing NNAs to support both the standard convolution and deformable convolution. Specifically, for the dynamic and irregular accesses in DCNs, we have both the input and output features divided into tiles and build a tile dependency table (TDT) to track the irregular tile dependency at runtime. With the TDT, we further develop an on-chip tile scheduler to handle the dynamic and irregular accesses efficiently. In addition, we propose a novel mapping strategy to enable parallel BLI processing on NNAs and apply layer fusion techniques for more energy-efficient DCN processing. According to our experiments, the proposed accelerator achieves orders of magnitude higher performance and energy efficiency compared to the typical computing architectures including ARM, ARM+TPU, and GPU with 6.6\% chip area penalty to a classic NNA.

cs.AR↗

arXiv2021

Shadow and Quasinormal Modes of a Rotating Loop Quantum Black Hole

In this paper, we construct an effective rotating loop quantum black hole (LQBH) solution, starting from the spherical symmetric LQBH by applying the Newman-Janis algorithm modified by Azreg-Aïnou's non-complexification procedure, and study the effects of loop quantum gravity { (LQG) on its shadow}. Given the rotating {LQBH}, we discuss its horizon, ergosurface, and regularity {as} $r \to 0$. Depending on the values of the specific angular momentum $a$ and the polymeric function $P$ arising from {LQG}, we {find} that the rotating solution we obtained can represent a regular black hole, a regular extreme black hole, or a regular spacetime {without horizon (a non-black-hole solution)}. We also {study} the effects of {LQG} and rotation, and {show} that, in addition to the specific angular momentum, the polymeric function {also} causes deformations in the size and shape of the black hole shadow. Interestingly, for a given value of $a$ and inclination angle $θ_0$, the apparent size of the shadow monotonically decreases, and the shadow gets more distorted with increasing $P$. We also {consider the effects of $P$ on the deviations from the circularity of the shadow, and find} that the deviation from circularity increases with increasing $P$ for fixed values of $a$ and $θ_0$. Additionally, we explore the observational implications of $P$ in comparison with the latest Event Horizon Telescope (EHT) observation of the supermassive black hole, M$87$. The connection between the shadow radius and quasinormal modes in the eikonal limit as well as {the} deflection of massive particles are also considered.

gr-qc↗

arXiv2021

Deciphering water-solid reactions during hydrothermal corrosion of SiC

Water solid interfacial reactions are critical to understanding corrosion. More specifically, it is notoriously difficult to determine how water and solid interact beyond the initial chemisorption to induce the surface dissolution. Here, we report atomic-scale mechanisms of the elementary steps during SiC hydrothermal corrosion, from the initial surface attack to surface dissolution. We find that hydrogen scission reactions play a vital role in breaking Si-C bonds, regardless of the surface orientations. Stable silica layer does not form on the surface, but the newly identified chemical reactions on SiC are analogous to those observed during the dissolution of silica. SiC is dissolved directly into the water as soluble silicic acid. The rate of hydrothermal corrosion determined based on the calculated reaction activation energies is consistent with available experimental data. Our work sheds new light on understanding and interpreting the experimental observations and it provides foundation for design of materials that are resistant to corrosion in water.

physics.chem-ph↗

arXiv2021

Effects of point defects on oxidation of 3C-SiC

The influence of implantation-induced point defects (PDs) on SiC oxidation is investigated via molecular dynamics simulations. PDs generally increase the oxidation rate of crystalline grains. Particularly, accelerations caused by Si antisites and vacancies are comparable, and followed by Si interstitials, which are higher than those by C antisites and C interstitials. However, in the grain boundary (GB) region, defect contribution to oxidation is more complex, with C antisites decelerating oxidation. The underlying reason is the formation of a C-rich region along the oxygen diffusion pathway that blocks the access of O to Si and thus reduces the oxidation rate, as compared to the oxidation along a GB without defects.

cond-mat.mtrl-sci↗

arXiv2021

An Unexpected Role of H During SiC Corrosion in Water

During aqueous corrosion, atoms in the solid react chemically with oxygen, leading either to the formation of an oxide film or to the dissolution of the host material. Commonly, the first step in corrosion involves an oxygen atom from the dissociated water that reacts with the surface atoms and breaks near surface bonds. In contrast, hydrogen on the surface often functions as a passivating species. Here, we discovered that the roles of O and H are reversed in the early corrosion stages on a Si terminated SiC surface. O forms stable species on the surface, and chemical attack occurs by H that breaks the Si-C bonds. This so-called hydrogen scission reaction is enabled by a newly discovered metastable bridging hydroxyl group that can form during water dissociation. The Si atom that is displaced from the surface during water attack subsequently forms H2SiO3, which is a known precursor to the formation of silica and silicic acid. This study suggests that the roles of H and O in oxidation need to be reconsidered.

cond-mat.mtrl-sci↗

arXiv2020

Thin Accretion Disk around a four-dimensional Einstein-Gauss-Bonnet Black Hole

Recently a novel four-dimensional Einstein-Gauss-Bonnet (4EGB) theory of gravity was proposed by Glavan and Lin [D. Glavan and C. Lin, Phys. Rev. Lett. 124, 081301 (2020)] which includes a regularized Gauss-Bonnet term by using the re-scaling of the Gauss-Bonnet coupling constant $α\to α/(D-4)$ in the limit $D\to 4$. This theory also has been reformulated to a specific class of the Horndeski theory with an additional scalar degree of freedom and to a spatial covariant version with a Lagrangian multiplier that can eliminate the scalar mode. Here we study the physical properties of the electromagnetic radiation emitted from a thin accretion disk around the static spherically symmetric black hole in the 4EGB gravity. For this purpose, we assume the disk is in a steady-state and in hydrodynamic and thermodynamic equilibrium so that the emitted electromagnetic radiation is a black body spectrum. We study in detail the effects of the Gauss-Bonnet coupling constant $α$ in 4EGB gravity on the energy flux, temperature distribution, and electromagnetic spectrum of the disk. It is shown that with the increases of the parameter $α$, the energy flux, temperature distribution, and electromagnetic spectrum of the accretion disk all increase. Besides, we also show that the accretion efficiency increases as the growth of the parameter $α$. Our results indicate that the thin accretion disk around the static spherically symmetric black hole in the 4EGB gravity is hotter, more luminosity, and more efficient than that around a Schwarzschild black hole with the same mass for a positive $α$, while it is cooler, less luminosity, and less efficient for a negative $α$.

gr-qc↗

arXiv2020

Evolution of magnetic order in van-der-Waals antiferromagnet FePS$_3$ through insulator-metal transition

Layered van-der-Waals 2D magnetic materials are of great interest in fundamental condensed-matter physics research, as well as for potential applications in spintronics and device physics. We present neutron powder diffraction data using new ultra-high-pressure techniques to measure the magnetic structure of Mott-insulating 2D honeycomb antiferromagnet FePS$_3$ at pressures up to 183 kbar and temperatures down to 80 K. These data are complemented by high-pressure magnetometry and reverse Monte Carlo modeling of the spin configurations. As pressure is applied, the previously-measured ambient-pressure magnetic order switches from an antiferromagnetic to a ferromagnetic interplanar interaction, and from 2D-like to 3D-like character. The overall antiferromagnetic structure within the $ab$ planes, ferromagnetic chains antiferromagnetically coupled, is preserved, but the magnetic propagation vector is altered from $(0\:1\:\frac{1}{2})$ to $(0\:1\:0)$, a halving of the magnetic unit cell size. At higher pressures, coincident with the second structural transition and the insulator-metal transition in this compound, we observe a suppression of this long-range-order and emergence of a form of magnetic short-range order which survives above room temperature. Reverse Monte Carlo fitting suggests this phase to be a short-ranged version of the original ambient pressure structure - with a return to antiferromagnetic interplanar correlations. The persistence of magnetism well into the HP-II metallic state is an observation in seeming contradiction with previous x-ray spectroscopy results which suggest a spin-crossover transition.

cond-mat.str-el↗

arXiv2020

Spherical Accretion Flow onto General Parameterized Spherically Symmetric Black Hole Spacetimes

The transonic phenomenon of black hole accretion and the existence of the photon sphere are the characteristics of strong gravitational fields near a black hole horizon. In this work, we study spherical accretion flow onto a general parametrized spherically symmetric black hole spacetimes. For this purpose, we analyze the accretion process of various perfect fluids, such as the isothermal fluid of ultra-stiff, ultra-relativistic, and sub-relativistic types and polytropic fluid, respectively. The influences of extra parameters beyond the Schwarzschild black hole in the general parameterized spherically symmetric black hole on the flow behaviors of the above-mentioned test fluids are studied in detail. In addition, by studying the accretion of ideal photon gas, we further discuss the correspondence between the sonic radius of accreting photon gas and the photon sphere for the general parameterized spherically symmetric black hole. Some possible future extensions of our analysis are also discussed.

gr-qc↗

arXiv2020

Structure and magnetism of a new hexagonal polymorph of Ba$_3$Tb(BO$_3$)$_3$ with a quasi-2D triangular lattice

This article reports the structural and magnetic characterisation of a new hexagonal ($P6_3cm$) low-temperature phase of Ba$_3$Tb(BO$_3$)$_3$, isostructural with the heavier lanthanide borates with formula Ba$_3$Ln(BO$_3$)$_3$ (Ln = Dy-Lu). The crystal structure contains a quasi-two-dimensional (2D) triangular lattice of Tb$^{3+}$ ions and is predicted to display unusual magnetic behaviour as a result of the low dimensionality. Magnetic susceptibility $χ(T)$ shows antiferromagnetic interactions ($θ_{CW} = -7.15$ K) and no sharp ordering transition at $T \geq 2$ K, but analysis of $χ'(T)$, isothermal magnetisation and heat capacity suggests the possibility of short-range ordering at $T \approx 10$ K. The material exhibits an inverse magnetocaloric effect at $T < 4$ K.

cond-mat.str-el↗

arXiv2020

Accelerating Generative Neural Networks on Unmodified Deep Learning Processors -- A Software Approach

Generative neural network is a new category of neural networks and it has been widely utilized in applications such as content generation, unsupervised learning, segmentation and pose estimation. It typically involves massive computing-intensive deconvolution operations that cannot be fitted to conventional neural network processors directly. However, prior works mainly investigated specialized hardware architectures through intensive hardware modifications to the existing deep learning processors to accelerate deconvolution together with the convolution. In contrast, this work proposes a novel deconvolution implementation with a software approach and enables fast and efficient deconvolution execution on the legacy deep learning processors. Our proposed method reorganizes the computation of deconvolution and allows the deep learning processors to treat it as the standard convolution by splitting the original deconvolution filters into multiple small filters. Compared to prior acceleration schemes, the implemented acceleration scheme achieves 2.41x - 4.34x performance speedup and reduces the energy consumption by 27.7% - 54.5% on a set of realistic benchmarks. In addition, we also applied the deconvolution computing approach to the off-the-shelf commodity deep learning processors. The performance of deconvolution also exhibits significant performance speedup over prior deconvolution implementations.

cs.LG↗

arXiv2020

EnGN: A High-Throughput and Energy-Efficient Accelerator for Large Graph Neural Networks

Graph neural networks (GNNs) emerge as a powerful approach to process non-euclidean data structures and have been proved powerful in various application domains such as social networks and e-commerce. While such graph data maintained in real-world systems can be extremely large and sparse, thus employing GNNs to deal with them requires substantial computational and memory overhead, which induces considerable energy and resource cost on CPUs and GPUs. In this work, we present a specialized accelerator architecture, EnGN, to enable high-throughput and energy-efficient processing of large-scale GNNs. The proposed EnGN is designed to accelerate the three key stages of GNN propagation, which is abstracted as common computing patterns shared by typical GNNs. To support the key stages simultaneously, we propose the ring-edge-reduce(RER) dataflow that tames the poor locality of sparsely-and-randomly connected vertices, and the RER PE-array to practice RER dataflow. In addition, we utilize a graph tiling strategy to fit large graphs into EnGN and make good use of the hierarchical on-chip buffers through adaptive computation reordering and tile scheduling. Overall, EnGN achieves performance speedup by 1802.9X, 19.75X, and 2.97X and energy efficiency by 1326.35X, 304.43X, and 6.2X on average compared to CPU, GPU, and a state-of-the-art GCN accelerator HyGCN, respectively.

cs.DC↗