arXiv subjects
Publications and source records attributed to Hao Jiang.
This report summarises our method and validation results for the ISIC Challenge 2018 - Skin Lesion Analysis Towards Melanoma Detection - Task 1: Lesion Segmentation. We present a two-stage method for lesion segmentation with optimised training method and ensemble post-process. Our method achieves state-of-the-art performance on lesion segmentation and we win the first place in ISIC 2018 task1.
Grant-free random access (RA) with massive MIMO is a promising RA technique with low signaling overhead that provides significant benefits in increasing the channel reuse efficiency. Since user equipment (UE) detection and channel estimation in grant-free RA rely solely on the received preambles, preamble designs that enable high success rate of UE detection and channel estimation are very much in need to ensure the performance gain of grant-free RA with massive MIMO. In this paper, a super preamble consisting of multiple consecutive preambles is proposed for high success rate of grant-free RA with massive MIMO. With the proposed approach, the success of UE detection and channel estimation for a RA UE depends on two conditions: 1) it is a solvable UE; 2) its super preamble is detected. Accordingly, we theoretically analyze the solvable rate of RA UEs with multiple preambles and propose a reliable UE detection algorithm to obtain the super preambles of RA UEs by exploiting the quasi-orthogonality characteristic of massive MIMO. Theoretical analysis and simulation results show that turning a preamble into a super preamble consisting of two or three shorter preambles, the success rate of UE detection and channel estimation could be significantly increased using the proposed approach.
We report the synthesis, crystal structure, physical properties, and first-principles calculations of a vanadium-based oxytelluride Rb$_{1-δ}$V$_2$Te$_2$O ($δ\approx0.2$). The crystal structure bears two-dimensional V$_2$O square nets sandwiched with tellurium layers, mimicking the structural motifs of cuprate and iron-based superconductors. The material exhibits metallic conductivity with dominant hole-type charge carriers. A weak metal-to-metal transition takes place at $\sim$100 K, which is conformably characterized by a slight kink/hump in the electrical resistivity, jumps in the Hall and Seebeck coefficients, a minute drop in the magnetic susceptibility, and a small peak in the heat capacity. Neither Bragg-peak splitting nor superlattice reflections can be detected within the resolution of conventional x-ray diffractions. The band-structure calculations show that V-3$d$ orbitals dominate the electronic states at around Fermi energy where a $d_{yz}/d_{xz}$ orbital polarization shows up. There are three Fermi-surface sheets that seem unfavorable for nesting. Our results suggest an orbital or spin-density-wave order for the low-temperature state and, upon suppression of the competing order, emergence of superconductivity could be expected.
The 5th generation mobile communication systems aim to support massive access for future wireless applications. Unfortunately, wireless resource scarcity in random access (RA) is a fundamental bottleneck for enabling massive access. To address this problem, we propose a virtual carrier sensing (VCS) based RA scheme in massive MIMO systems. The essence of the proposed scheme lies in exploiting wireless spatial resources of uplink channels occupied by assigned user equipments (UEs) to increase channel resources for RA. With the proposed scheme, RA UEs are able to exploit the spatial resources that are approximately orthogonal to those of assigned UEs, thus sharing the uplink channel resource with assigned UEs without causing significant interference to them. Specifically, to ensure RA UEs avoid serious interference with assigned UEs, base station (BS) sends tailored virtual carriers to RA UEs on behalf of assigned UEs. RA UEs then conduct VCS to determine whether or not the uplink channel resource is available for RA. Closed-form approximations for probability of channel availability and uplink achievable rate with the proposed scheme are derived. Theoretical analysis and simulation results show that the proposed scheme is able to significantly increase channel resources of RA for massive access.
Sr2Cr3As2O2 is composed of alternating square-lattice CrO2 and Cr2As2 stacking layers, where CrO2 is isostructural to the CuO2 building-block of cuprate high-Tc superconductors and Cr2As2 to Fe2As2 of Fe-based superconductors. Current interest in this material is raised by theoretic prediction of possible superconductivity. In this neutron powder diffraction study, we discovered that magnetic moments of Cr(II) ions in the Cr2As2 sublattice develop a C-type antiferromagnetic structure below 590 K, and the moments of Cr(I) in the CrO2 sublattice form the La2CuO4 -like antiferromagnetic order below 291 K. The staggered magnetic moment 2.19(4)μ B /Cr(II) in the more itinerant Cr2As2 layer is smaller than 3.10(6)μ_B/Cr(I) in the more localized CrO2 layer. Different from previous expectation, a spin-flop transition of the Cr(II) magnetic order observed at 291 K indicates a strong coupling between the CrO2 and Cr2As2 magnetic subsystems.
Recognizing every person's action in a crowded and cluttered environment is a challenging task. In this paper, we propose a real-time action recognition method, Action4D, which gives reliable and accurate results in the real-world settings. We propose to tackle the action recognition problem using a holistic 4D "scan" of a cluttered scene to include every detail about the people and environment. Recognizing multiple people's actions in the cluttered 4D representation is a new problem. In this paper, we propose novel methods to solve this problem. We propose a new method to track people in 4D, which can reliably detect and follow each person in real time. We propose a new deep neural network, the Action4D-Net, to recognize the action of each tracked person. The Action4D-Net's novel structure uses both the global feature and the focused attention to achieve state-of-the-art result. Our real-time method is invariant to camera view angles, resistant to clutter and able to handle crowd. The experimental results show that the proposed method is fast, reliable and accurate. Our method paves the way to action recognition in the real-world applications and is ready to be deployed to enable smart homes, smart factories and smart stores.
Recent breakthroughs in recurrent deep neural networks with long short-term memory (LSTM) units has led to major advances in artificial intelligence. State-of-the-art LSTM models with significantly increased complexity and a large number of parameters, however, have a bottleneck in computing power resulting from limited memory capacity and data communication bandwidth. Here we demonstrate experimentally that LSTM can be implemented with a memristor crossbar, which has a small circuit footprint to store a large number of parameters and in-memory computing capability that circumvents the 'von Neumann bottleneck'. We illustrate the capability of our system by solving real-world problems in regression and classification, which shows that memristor LSTM is a promising low-power and low-latency hardware platform for edge inference.
Memristor is a promising building block for the next generation nonvolatile random access memory and bio-inspired computing systems. Organizing memristors into high density crossbar arrays, although challenging, is critical to meet the ever-growing high capacity and low energy demands of these applications especially in the big data era. Here, we construct memristor crossbars with a single-layer density up to 4.5 terabits per inch square, an order of magnitude denser than the state- of-the-art 64-layer triple level cell NAND flash technology. The memristors in the crossbars are 2 $\times$ 2 nm$^2$ in size, capable of switching with tens of nano ampere electric current. The densely packed memristor crossbars of extremely small working devices provides a power-efficient solution for high density information storage and processing.
Massive MIMO opens up new avenues for enabling highly efficient random access (RA) by offering abundance of spatial degrees of freedom. In this paper, we investigate the grant-free RA with massive MIMO and derive the analytic expressions of success probability of the grant-free RA for conjugate beamforming and zero-forcing beamforming techniques.With the derived analytic expressions, we further shed light on the impact of system parameters on the success probability. Simulation results verify the accuracy of the analyses. It is confirmed that the grant-free RA with massive MIMO is an attractive RA technique with low signaling overhead that could simultaneously accommodate a number of RA users, which is multiple times the number of RA channels, with close-to-one success probability. In addition, when the number of antennas in massive MIMO is sufficiently large, we show that the number of orthogonal preambles would dominate the success probability.
In this paper, we consider solving a class of nonconvex and nonsmooth problems frequently appearing in signal processing and machine learning research. The traditional alternating direction method of multipliers encounters troubles in both mathematics and computations in solving the nonconvex and nonsmooth subproblem. In view of this, we propose a reweighted alternating direction method of multipliers. In this algorithm, all subproblems are convex and easy to solve. We also provide several guarantees for the convergence and prove that the algorithm globally converges to a critical point of an auxiliary function with the help of the Kurdyka-Łojasiewicz property. Several numerical results are presented to demonstrate the efficiency of the proposed algorithm.
The eigenvalue of the hermitic Hamiltonian is real undoubtedly. Actually, The reality can also be guaranteed by the $PT$-symmetry. The hermiticity and the $PT$-symmetric quantum theory both have requirements regarding the boundary condition. There exists a reverse strategy to investigate the quantum problem. Namely, define the eigenvalue as real first, and, meanwhile, open the boundary condition. Then the behaviors of the wave function at the boundary become rich in meaning. This eigenfunction is generally divergent, and the extent and direction of divergence are closely linked to the energy. It was noted that these divergent behaviors can be well described by their energy-space uncertainty relation which is not trivial anymore. The divergent state is unstable and will certainly exchange energy with the outside. The mechanism of energy exchange is just in the energy-space uncertainty relation, which will benefit dynamic simulation, the many-body problem, and so on. There is no distinct dividing line between this kind of divergent unstable state and the convergent stable state. Their relationship is like that of the rational and irrational numbers. In practice, there are distinct advantages of speed and accuracy for the methods based on the laws of divergence.
The high incidence rate of prostate disease poses a requirement in early detection for diagnosis. As one of the main imaging methods used for prostate cancer detection, Magnetic Resonance Imaging (MRI) has wide range of appearance and imbalance problems, making automated prostate segmentation fundamental but challenging. Here we propose a novel Densely Dilated Spatial Pooling Convolutional Network (DDSP ConNet) in encoder-decoder structure. It employs dense structure to combine dilated convolution and global pooling, thus supplies coarse segmentation results from encoder and decoder subnet and preserves more contextual information. To obtain richer hierarchical feature maps, residual long connection is furtherly adopted to fuse contexture features. Meanwhile, we adopt DSC loss and Jaccard loss functions to train our DDSP ConNet. We surprisingly found and proved that, in contrast to re-weighted cross entropy, DSC loss and Jaccard loss have a lot of benign properties in theory, including symmetry, continuity and differentiability about the parameters of network. Extensive experiments on the MICCAI PROMISE12 challenge dataset have been done to corroborate the effectiveness of our DDSP ConNet with DSC loss and Jaccard loss. Totally, our method achieves a score of 85.78 in the test dataset, outperforming most of other competitors.
In this paper, we present a three-dimensional (3D) non-wide-sense stationary (non-WSS) wideband geometry-based channel model for vehicle-to-vehicle (V2V) communication environments. We introduce a two-cylinder model to describe moving vehicles as well as multiple confocal semi-ellipsoid models to depict stationary roadside scenarios. The received signal is constructed as a sum of the line-of-sight (LoS), single-, and double-bounced rays with different energies. Accordingly, the proposed channel model is sufficient for depicting a wide variety of V2V environments, such as macro-, micro-, and picocells. The relative movement between the mobile transmitter (MT) and mobile receiver (MR) results in time-variant geometric statistics that make our channel model non-stationary. Using this channel model, the proposed channel statistics, i.e., the time-variant space correlation functions (CFs), frequency CFs, and corresponding Doppler power spectral density (PSD), were studied for different relative moving time instants. The numerical results demonstrate that the proposed 3D non-WSS wideband channel model is practical for characterizing real V2V channels.
We used molecular dynamics simulations and the path sampling technique known as forward flux sampling to study homogeneous nucleation of NaCl crystals from supersaturated aqueous solutions at 298 K and 1 bar. Nucleation rates were obtained for a range of salt concentrations for the Joung-Cheatham NaCl force field combined with the SPC/E water model. The calculated nucleation rates are significantly lower than available experimental measurements. The estimates for the nucleation rates in this work do not rely on classical nucleation theory, but the pathways observed in the simulations suggest that the nucleation process is better described by classical nucleation theory than an alternative interpretation based on Ostwald's step rule, in contrast to some prior simulations of related models. In addition to the size of NaCl nucleus, we find that the crystallinity of a nascent cluster plays an important role in the nucleation process. Nuclei with high crystallinity were found to have higher growth probability and longer lifetimes, possibly because they are less exposed to hydration water.
The distributed computation of Nash equilibria is assuming growing relevance in engineering where such problems emerge in the context of distributed control. Accordingly, we present schemes for computing equilibria of two classes of static stochastic convex games complicated by a parametric misspecification, a natural concern in the control of large-scale networked engineered system. In both schemes, players learn the equilibrium strategy while resolving the misspecification: (1) Monotone stochastic Nash games: We present a set of coupled stochastic approximation schemes distributed across agents in which the first scheme updates each agent's strategy via a projected (stochastic) gradient step while the second scheme updates every agent's belief regarding its misspecified parameter using an independently specified learning problem. We proceed to show that the produced sequences converge in an almost-sure sense to the true equilibrium strategy and the true parameter, respectively. Surprisingly, convergence in the equilibrium strategy achieves the optimal rate of convergence in a mean-squared sense with a quantifiable degradation in the rate constant; (2) Stochastic Nash-Cournot games with unobservable aggregate output: We refine (1) to a Cournot setting where we assume that the tuple of strategies is unobservable while payoff functions and strategy sets are public knowledge through a common knowledge assumption. By utilizing observations of noise-corrupted prices, iterative fixed-point schemes are developed, allowing for simultaneously learning the equilibrium strategies and the misspecified parameter in an almost-sure sense.
A recent version of statistical associating fluid theory (SAFT), namely SAFT2, is coupled with the van der Waals and Platteeuw theory to study the alkane hydrate phase equilibrium conditions. The model is found to provide an accurate representation of the alkane hydrate dissociation conditions with and without inhibitors, such as salts, alcohols, as well as mixed salts and alcohol. Based on SAFT2, a heterosegmented SAFT equation of state is developed to model the thermodynamic properties of aqueous ionic liquid (IL) solutions, which is recently discovered as dual function gas hydrate inhibitors. With transferrable model parameters, the heterosegmented SAFT generally well represents the liquid density, activity coefficient, and osmotic coefficient of aqueous imidazolium IL solutions. The inhibition effects of imidazolium IL on methane hydrate is also studied by the heterosegmented SAFT and the van der Waals and Platteeuw theory. The heterosegmented SAFT is then modified to better represent the thermodynamic properties of aqueous IL solutions with the help of Monte Carlo simulation and the solutions of Ornstein-Zernike integral equation. A simple modification of MSA, referred to as KMSA, is proposed to accurately predict the excess energies of electrolyte system in mixture with neutral component. Monte Carlo simulations are also conducted on flexible charged hard-sphere chain molecules, and a SAFT model which implements either a dimer or a dimer-monomer approach to account for the charged chain connectivity is proposed.
In view of the fact that biological characteristics have excellent independent distinguishing characteristics,biometric identification technology involves almost all the relevant areas of human distinction. Fingerprints, iris, face, voice-print and other biological features have been widely used in the public security departments to detect detection, mobile equipment unlock, target tracking and other fields. With the use of electronic devices more and more widely and the frequency is getting higher and higher. Only the Biometrics identification technology with excellent recognition rate can guarantee the long-term development of these fields.
We report the first result on Ge-76 neutrinoless double beta decay from CDEX-1 experiment at China Jinping Underground Laboratory. A mass of 994 g p-type point-contact high purity germanium detector has been installed to search the neutrinoless double beta decay events, as well as to directly detect dark matter particles. An exposure of 304 kg*day has been analyzed. The wideband spectrum from 500 keV to 3 MeV was obtained and the average event rate at the 2.039 MeV energy range is about 0.012 count per keV per kg per day. The half-life of Ge-76 neutrinoless double beta decay has been derived based on this result as: T 1/2 > 6.4*10^22 yr (90% C.L.). An upper limit on the effective Majorana-neutrino mass of 5.0 eV has been achieved. The possible methods to further decrease the background level have been discussed and will be pursued in the next stage of CDEX experiment.