arXiv subjects

Kai He

Publications and source records attributed to Kai He.

71 records · Page 4Linked to original sources

arXiv2022

Theoretical and experimental study on Noise Equivalent Power of X-ray semiconductor ultra-fast response material based on the rad-optic effect

Semiconductor material based on the rad-optic effect enables ultra-fast detection of X-rays and plays an important role in fusion diagnostics. Obtaining the accurate noise equivalent power (NEP) of the semiconductor ultrafast response material is the key to detecting X-rays. In this paper, the refractive index change mechanism of the semiconductor under X-ray irradiation was analyzed, and the quantitative relationship between the diffraction efficiency and the X-ray photon energy was established through the LT-AlGaAs diffraction imaging experiments. The impulse responses of LT-AlGaAs under 1 KeV-10 KeV X-ray radiation were calculated, revealing the variation of NEP density with radiated photon energy. In the case of bombarding the Al target to generate 1.5 KeV X-rays, the imaging experiments of LT-AlGaAs were performed. The diffraction image of LT-AlGaAs has a linear relationship with the radiation intensity, and the NEP density of LT-AlGaAs reaches 4.80*105W/cm2. This study has reference significance for the development of ultra-fast X-ray imaging systems based on the rad-optic effect.

physics.ins-det↗

arXiv2022

Learning Interpretable Decision Rule Sets: A Submodular Optimization Approach

Rule sets are highly interpretable logical models in which the predicates for decision are expressed in disjunctive normal form (DNF, OR-of-ANDs), or, equivalently, the overall model comprises an unordered collection of if-then decision rules. In this paper, we consider a submodular optimization based approach for learning rule sets. The learning problem is framed as a subset selection task in which a subset of all possible rules needs to be selected to form an accurate and interpretable rule set. We employ an objective function that exhibits submodularity and thus is amenable to submodular optimization techniques. To overcome the difficulty arose from dealing with the exponential-sized ground set of rules, the subproblem of searching a rule is casted as another subset selection task that asks for a subset of features. We show it is possible to write the induced objective function for the subproblem as a difference of two submodular (DS) functions to make it approximately solvable by DS optimization algorithms. Overall, the proposed approach is simple, scalable, and likely to be benefited from further research on submodular optimization. Experiments on real datasets demonstrate the effectiveness of our method.

cs.LG↗

arXiv2022

GenISP: Neural ISP for Low-Light Machine Cognition

Object detection in low-light conditions remains a challenging but important problem with many practical implications. Some recent works show that, in low-light conditions, object detectors using raw image data are more robust than detectors using image data processed by a traditional ISP pipeline. To improve detection performance in low-light conditions, one can fine-tune the detector to use raw image data or use a dedicated low-light neural pipeline trained with paired low- and normal-light data to restore and enhance the image. However, different camera sensors have different spectral sensitivity and learning-based models using raw images process data in the sensor-specific color space. Thus, once trained, they do not guarantee generalization to other camera sensors. We propose to improve generalization to unseen camera sensors by implementing a minimal neural ISP pipeline for machine cognition, named GenISP, that explicitly incorporates Color Space Transformation to a device-independent color space. We also propose a two-stage color processing implemented by two image-to-parameter modules that take down-sized image as input and regress global color correction parameters. Moreover, we propose to train our proposed GenISP under the guidance of a pre-trained object detector and avoid making assumptions about perceptual quality of the image, but rather optimize the image representation for machine cognition. At the inference stage, GenISP can be paired with any object detector. We perform extensive experiments to compare our method to other low-light image restoration and enhancement methods in an extrinsic task-based evaluation and validate that GenISP can generalize to unseen sensors and object detectors. Finally, we contribute a low-light dataset of 7K raw images annotated with 46K bounding boxes for task-based benchmarking of future low-light image restoration and object detection.

cs.CV↗

arXiv2022

NetRCA: An Effective Network Fault Cause Localization Algorithm

Localizing the root cause of network faults is crucial to network operation and maintenance. However, due to the complicated network architectures and wireless environments, as well as limited labeled data, accurately localizing the true root cause is challenging. In this paper, we propose a novel algorithm named NetRCA to deal with this problem. Firstly, we extract effective derived features from the original raw data by considering temporal, directional, attribution, and interaction characteristics. Secondly, we adopt multivariate time series similarity and label propagation to generate new training data from both labeled and unlabeled data to overcome the lack of labeled samples. Thirdly, we design an ensemble model which combines XGBoost, rule set learning, attribution model, and graph algorithm, to fully utilize all data information and enhance performance. Finally, experiments and analysis are conducted on the real-world dataset from ICASSP 2022 AIOps Challenge to demonstrate the superiority and effectiveness of our approach.

cs.LG↗

arXiv2021

RobustPeriod: Time-Frequency Mining for Robust Multiple Periodicity Detection

Periodicity detection is a crucial step in time series tasks, including monitoring and forecasting of metrics in many areas, such as IoT applications and self-driving database management system. In many of these applications, multiple periodic components exist and are often interlaced with each other. Such dynamic and complicated periodic patterns make the accurate periodicity detection difficult. In addition, other components in the time series, such as trend, outliers and noises, also pose additional challenges for accurate periodicity detection. In this paper, we propose a robust and general framework for multiple periodicity detection. Our algorithm applies maximal overlap discrete wavelet transform to transform the time series into multiple temporal-frequency scales such that different periodic components can be isolated. We rank them by wavelet variance, and then at each scale detect single periodicity by our proposed Huber-periodogram and Huber-ACF robustly. We rigorously prove the theoretical properties of Huber-periodogram and justify the use of Fisher's test on Huber-periodogram for periodicity detection. To further refine the detected periods, we compute unbiased autocorrelation function based on Wiener-Khinchin theorem from Huber-periodogram for improved robustness and efficiency. Experiments on synthetic and real-world datasets show that our algorithm outperforms other popular ones for both single and multiple periodicity detection.

cs.LG↗

arXiv2020

The Detection of Thoracic Abnormalities ChestX-Det10 Challenge Results

The detection of thoracic abnormalities challenge is organized by the Deepwise AI Lab. The challenge is divided into two rounds. In this paper, we present the results of 6 teams which reach the second round. The challenge adopts the ChestX-Det10 dateset proposed by the Deepwise AI Lab. ChestX-Det10 is the first chest X-Ray dataset with instance-level annotations, including 10 categories of disease/abnormality of 3,543 images. The annotations are located at https://github.com/Deepwise-AILab/ChestX-Det10-Dataset. In the challenge, we randomly split all data into 3001 images for training and 542 images for testing.

eess.IV↗

arXiv2020

Grain-boundary structure and segregation in Nb3Sn coatings on Nb for high-performance superconducting radiofrequency cavity applications

We report on atomic-scale analyses of grain boundary (GB) structures and segregation in Nb3Sn coatings on Nb, prepared by the vapor-diffusion process, for superconducting radiofrequency (SRF) cavity applications, utilizing atom-probe tomography, high-resolution scanning transmission electron-microscopy and first-principles calculations. We demonstrate that the chemical composition of Nb3Sn GBs is correlated strongly with the diffusion of Sn and Nb at GBs during the coating process. In a sample coated with a relatively large Sn flux, we observe an interfacial width of Sn segregation at a GB of ~3 nm, with a maximum concentration of ~35 at.%. After post-annealing at 1100 oC for 3 h, the Sn segregated at GBs disappears and Nb segregation is observed subsequently at GBs, indicating that Nb diffused into the Nb3Sn GBs from the Nb substrate. It is also demonstrated that the amount of Sn segregation in a Nb3Sn coating can be controlled by: (i) Sn flux; and (ii) the temperatures of the Nb substrates and Sn source, which may affect the overall kinetics including GB diffusion of Sn and Nb. An investigation of the correlation between the chemical compositions of GBs and Nb3Sn SRF cavity performance reveals that the Nb3Sn SRF cavities with the best performance (high-quality factors at high accelerating electric-field gradients) do not exhibit Sn segregation at GBs. Our results suggest that the chemical compositions of GBs in Nb3Sn coatings for SRF cavities can be controlled by GB engineering and can be utilized to optimize fabrication of high-quality Nb3Sn coatings for SRF cavities.

cond-mat.mtrl-sci↗

arXiv2019

Atomic scale analysis of Nb3Sn on Nb prepared by a vapor-diffusion process for superconducting radiofrequency cavity applications

We report an atomic-scale analysis of the microstructure of Nb3Sn coating on Nb prepared by vapor diffusion process for superconducting radiofrequency (SRF) cavity application using transmission electron microscopy (TEM). Epitaxial growth of Nb3Sn on the Nb substrate is found and four types of orientation relationships at the Nb3Sn/Nb interface are identified by electron diffraction or high-resolution scanning transmission electron microscopy (STEM) analysis. Thin Nb3Sn grains are observed in regions with low Sn flux and they have the specific orientation relationship, Nb3Sn (1-20)//Nb (-111) and Nb3Sn (002)//Nb (0-11). The Nb3Sn/Nb interface of thin grains had a large lattice mismatch, 12.3 at.%, and a high density of misfit dislocations was observed by HR-STEM. Based on our microstructural analysis of the thin grains, we conclude that the thin regions are probably a result of a slow interfacial reaction with this particular orientation relationship at the interface. The Sn-deficient regions are seen to form initially at the Nb3Sn/Nb interface and remain in the grains due to the slow diffusion of Sn in bulk Nb3Sn. The formation of Sn-deficient regions and the effects of strain and interfacial energies on the formation of Sn-deficient regions at various interfaces were also estimated by first-principle calculation. The finding of orientation relationships at the Nb3Sn/Nb interface provides important information about the formation of defects in Nb3Sn coatings such as large thin regions, Sn-deficient regions, which are critical to the performance of Nb3Sn superconducting radiofrequency cavities for accelerators.

cond-mat.mtrl-sci↗

arXiv2019

Through-the-Wall Imaging Exploiting 2.4GHz Commodity Wi-Fi

In this letter, we experimentally investigate a low-cost through-the-wall imaging exploiting Wi-Fi signals in an indoor environment from the perspective of holographic imaging. In our experiments, a pair of antennas in a synthetic aperture mode is used to acquire signals produced by commodity Wi-Fi devices and reflected from the scene in a synthetic aperture mode. The classical filtered back propagation (FBP) algorithm is then employed to form the image based on these signals. We use an IEEE 802.11n wireless router working at 2.4GHz with bandwidth of 20MHz. Selected experimental results are provided to demonstrate the performance of the proposed Wi-Fi based imaging scheme.

eess.SP↗

arXiv2016

Degradation of Black Phosphorus (BP): The Role of Oxygen and Water

Black phosphorus (BP) has attracted significant interest as a monolayer or few-layer material with extraordinary electrical and optoelectronic properties. However, degradation in air and other environments is an unresolved issue that may limit future applications. In particular the role of different ambient species has remained controversial. Here, we report systematic experiments combined with ab-initio calculations that address the effects of oxygen and water in the degradation of BP. Our results show that BP rapidly degrades whenever oxygen is present, but is unaffected by deaerated (i.e., O2 depleted) water. This behavior is rationalized by oxidation involving a facile dissociative chemisorption of O2, whereas H2O molecules are weakly physisorbed and do not dissociate on the BP surface. Oxidation (by O2) turns the hydrophobic pristine BP surface progressively hydrophilic. Our results have implications on the development of encapsulation strategies for BP, and open new avenues for exploration of phenomena in aqueous solutions including solution-gating, electrochemistry, and solution-phase approaches for exfoliation, dispersion, and delivery of BP.

cond-mat.mtrl-sci↗

arXiv2014

Driven dipole oscillations and the lowest energy excitations of strongly interacting lattice bosons in a harmonic trap

We show that the analysis of the time evolution of the occupation of site and momentum modes of harmonically trapped lattice hard-core bosons, under driven dipole oscillations, allows one to determine the energy of the lowest one-particle excitations of the system in equilibrium. The analytic solution of a single particle in the absence of a lattice is used to identify which function of those time-dependent observables is best fit for the analysis, as well as to relate the dynamic response of the system to its single-particle spectrum. In the presence of the lattice and of multiple particles, a much richer and informative dynamical response is observed under the drive.

cond-mat.quant-gas↗

arXiv2014

Localization of Wiener Functionals of Fractional Regularity and Applications

In this paper we localize some of Watanabe's results on fractional Wiener functionals, and use them to give a precise estimate of the difference between two Donsker's delta functionals even with fractional differentiability. As an application, the convergence rate of the density of the Euler scheme for non-Markovian stochastic differential equations is obtained.

math.PR↗

arXiv2013

Single-particle and many-body analyses of a quasiperiodic integrable system after a quench

In general, isolated integrable quantum systems have been found to relax to an apparent equilibrium state in which the expectation values of few-body observables are described by the generalized Gibbs ensemble. However, recent work has shown that relaxation to such a generalized statistical ensemble can be precluded by localization in a quasiperiodic lattice system. Here we undertake complementary single-particle and many-body analyses of noninteracting spinless fermions and hard-core bosons within the Aubry-Andre model to gain insight into this phenomenon. Our investigations span both the localized and delocalized regimes of the quasiperiodic system, as well as the critical point separating the two. Considering first the case of spinless fermions, we study the dynamics of the momentum distribution function and characterize the effects of real-space and momentum-space localization on the relevant single-particle wave functions and correlation functions. We show that although some observables do not relax in the delocalized and localized regimes, the observables that do relax in these regimes do so in a manner consistent with a recently proposed Gaussian equilibration scenario, whereas relaxation at the critical point has a more exotic character. We also construct various statistical ensembles from the many-body eigenstates of the fermionic and bosonic Hamiltonians and study the effect of localization on their properties.

cond-mat.quant-gas↗

arXiv2013

Initial state dependence of the quench dynamics in integrable quantum systems. III. Chaotic states

We study sudden quantum quenches in which the initial states are selected to be either eigenstates of an integrable Hamiltonian that is nonmappable to a noninteracting one or a nonintegrable Hamiltonian, while the Hamiltonian after the quench is always integrable and mappable to a noninteracting one. By studying weighted energy densities and entropies, we show that quenches starting from nonintegrable (chaotic) eigenstates lead to an "ergodic" sampling of the eigenstates of the final Hamiltonian, while those starting from the integrable eigenstates do not (or at least it is not apparent for the system sizes accessible to us). This goes in parallel with the fact that the distribution of conserved quantities in the initial states is thermal in the nonintegrable cases and nonthermal in the integrable ones, and means that, in general, thermalization occurs in integrable systems when the quench starts form an eigenstate of a nonintegrable Hamiltonian (away from the edges of the spectrum), while it fails (or requires larger system sizes than those studied here to become apparent) for quenches starting at integrable points. We test those conclusions by studying the momentum distribution function of hard-core bosons after a quench.

cond-mat.stat-mech↗

arXiv2012

Noise Correlation Scalings: Revisiting the Quantum Phase Transition in Incommensurate Lattices with Hard-Core Bosons

Finite size scalings of the momentum distribution and noise correlations are performed to study Mott insulator, Bose glass, and superfluid quantum phases in hard-core bosons (HCBs) subjected to quasi-periodic disorder. The exponents of the correlation functions at the Superfluid to Bose glass (SF-BG) transition are found to be approximately one half of the ones that characterizes the superfluid phase. The derivatives of the peak intensities of the correlation functions with respect to quasiperiodic disorder are shown to diverge at the SF-BG critical point. This behavior does not occur in the corresponding free fermion system, which also exhibits an Anderson-like transition at the same critical point, and thus provides a unique experimental tool to locate the phase transition in interacting bosonic systems. We also report on the absence of primary sublattice peaks in the momentum distribution of the superfluid phase for special fillings.

cond-mat.quant-gas↗

arXiv2012

Initial state dependence of the quench dynamics in integrable quantum systems. II. Thermal states

We study properties of isolated integrable quantum systems after a sudden quench starting from thermal states. We show that, even if the system is initially in thermal equilibrium at finite temperature, the diagonal entropy after a quench remains a fraction of the entropy in the generalized ensembles introduced to describe integrable systems after relaxation. The latter is also, in general, different from the entropy in thermal equilibrium. Furthermore, we examine the difference between the distribution of conserved quantities in the thermal and generalized ensembles after a quench and show that they are also, in general, different from each other. This explains why these systems fail to thermalize. A finite size scaling analysis is presented for each quantity, which allows us making predictions for thermodynamically large lattice sizes.

cond-mat.quant-gas↗

arXiv2011

Scaling of noise correlations in one-dimensional-lattice-hard-core-boson systems

Noise correlations are studied for systems of hard-core bosons in one-dimensional lattices. We use an exact numerical approach based on the Bose-Fermi mapping and properties of Slater determinants. We focus on the scaling of the noise correlations with system size in superfluid and insulating phases, which are generated in the homogeneous lattice, with period-two superlattices and with uniformly distributed random diagonal disorder. For the superfluid phases, the leading contribution is shown to exhibit a density-independent scaling proportional to the system size, while the first subleading term exhibits a density-dependent power-law exponent.

cond-mat.quant-gas↗