arXiv subjects
Publications and source records attributed to Zhipeng Liu.
We consider the geodesic of the directed last passage percolation with iid exponential weights. We find the explicit one-point distribution of the geodesic location joint with the last passage times, and its limit as the parameters go to infinity under the KPZ scaling.
The parabolic Airy process is the Airy$_2$ process minus a parabola, initially defined by its finite-dimensional distributions, which are given by a Fredholm determinant formula with the extended Airy kernel. This process is also the one-time spatial marginal of the KPZ fixed point with the narrow wedge initial condition. There are two formulas for the space-time multipoint distribution of the KPZ fixed point with the narrow wedge initial condition obtained by arXiv:1906.01053 and arXiv:1907.09876. Especially, the equal-time case of arXiv:1907.09876 gives a different formula of the multipoint distribution of the parabolic Airy process. In this paper, we present a direct proof that this formula matches the one with the extended Airy kernel. Some byproducts in the proof include several new formulas for the parabolic Airy process, and a generalization of the Andreief's identity.
Real-world time series typically exhibit complex temporal variations, making the time series classification task notably challenging. Recent advancements have demonstrated the potential of multi-scale analysis approaches, which provide an effective solution for capturing these complex temporal patterns. However, existing multi-scale analysis-based time series prediction methods fail to eliminate redundant scale-shared features across multi-scale time series, resulting in the model over- or under-focusing on scale-shared features. To address this issue, we propose a novel end-to-end Disentangled Multi-Scale framework for Time Series classification (DisMS-TS). The core idea of DisMS-TS is to eliminate redundant shared features in multi-scale time series, thereby improving prediction performance. Specifically, we propose a temporal disentanglement module to capture scale-shared and scale-specific temporal representations, respectively. Subsequently, to effectively learn both scale-shared and scale-specific temporal representations, we introduce two regularization terms that ensure the consistency of scale-shared representations and the disparity of scale-specific representations across all temporal scales. Extensive experiments conducted on multiple datasets validate the superiority of DisMS-TS over its competitive baselines, with the accuracy improvement up to 9.71%.
The KPZ fixed point is a (1+1)-dimensional space-time random field conjectured to be the universal limit for models within the Kardar-Parisi-Zhang (KPZ) universality class. We consider the KPZ fixed point with the narrow-wedge initial condition, conditioning on a large value at a specific point. By zooming in the neighborhood of this high point appropriately, we obtain a limiting random field, which we call an upper tail field of the KPZ fixed point. Different from the KPZ fixed point, where the time parameter has to be nonnegative, the upper tail field is defined in the full $2$-dimensional space. Especially, if we zoom out the upper tail field appropriately, it behaves like a Brownian-type field in the negative time regime, and the KPZ fixed point in the positive time regime. One main ingredient of the proof is an upper tail estimate of the joint tail probability functions of the KPZ fixed point near the given point, which generalizes the well known one-point upper tail estimate of the GUE Tracy-Widom distribution.
The periodic KPZ fixed point is the conjectural universal limit of the KPZ universality class models on a ring when both the period and time critically tend to infinity. For the case of the periodic narrow wedge initial condition, we consider the conditional distribution when the periodic KPZ fixed point is unusually large at a particular position and time. We prove a conditional limit theorem up to the ``pinch-up" time. When the period is large enough, the result is the same as that for the KPZ fixed point on the line obtained by Liu and Wang in 2022. We identify the regimes in which the result changes and find probabilistic descriptions of the limits.
This study investigates the droplet coalescence mechanisms and the interplay between various thermodynamic non-equilibrium (TNE) effects under isothermal and non-isothermal conditions kinetically. The main findings include: (1) Coalescence initiation and cut-through mechanisms: In non-isothermal conditions, the temperature rise caused by the release of latent heat during phase transition slightly increases the surface tension gradient (driving force) near the contact point of the two droplets, while significantly enhancing the pressure gradient (resistance). This results in a significantly prolonged coalescence initiation time compared to the isothermal case. In both cases, pressure extends the liquid-vapor interface in opposite directions, promoting the growth of the liquid bridge. (2) TNE effects: Latent heat-induced temperature rise significantly refrains the TNE intensity in thermal case. Before and after droplet contact, non-equilibrium quantities driven by the temperature gradient and those driven by the velocity gradient, alternate in dominating the coalescence process. This competition and interplay result in a more complex spatial and spatiotemporal evolution of TNE effects compared to the isothermal case. (3) Entropy production mechanisms: In the non-isothermal case, entropy production is contributed not only by $\bm Δ^{\ast}_2$ but also by $\bm Δ^{\ast}_{3,1}$, with the former being the dominant contributor. The temperature field reduces the entropy production rate, while extends its duration, and increases the total entropy production. This research provides kinetic insights for dynamic, cross-scale regulation and multifunctional integration of coalescence processes in industrial applications.
Before solar eruptions, a short-term slow-rise phase is often observed, during which the pre-eruption structure ascends at speeds much greater than the photospheric motions but much less than those of the eruption phase. Numerical magnetohydrodynamic (MHD) simulations of the coronal evolution driven by photospheric motions up to eruptions have been used to explain the slow-rise phase, but their bottom driving speeds are much larger than realistic photospheric values. Therefore, it remains an open question how the excessively fast bottom driving impacts the slow-rise phase. Here we modelled the slow-rise phase before eruption initiated from a continuously sheared magnetic arcade. In particular, we performed a series of experiments with the bottom driving speed unprecedentedly approaching the photospheric value of around $1$ km s$^{-1}$. The simulations confirmed that the slow-rise phase is an ideal MHD process, i.e., a manifestation of the growing expansion of the sheared arcade in the process of approaching a fully open field state. The overlying field line above the core flux has a slow-rise speed modulated by the driving speed's magnitude but is always over an order of magnitude larger than the driving speed. The core field also expands with speed much higher than the driving speed but much lower than that of the overlying field. By incrementally reducing the bottom-driving speed to realistic photospheric values, we anticipate better matches between the simulated slow-rise speeds and some observed ones.
Stock trend prediction involves forecasting the future price movements by analyzing historical data and various market indicators. With the advancement of machine learning, graph neural networks (GNNs) have been extensively employed in stock prediction due to their powerful capability to capture spatiotemporal dependencies of stocks. However, despite the efforts of various GNN stock predictors to enhance predictive performance, the improvements remain limited, as they focus solely on analyzing historical spatiotemporal dependencies, overlooking the correlation between historical and future patterns. In this study, we propose a novel distillation-based future-aware GNN framework (DishFT-GNN) for stock trend prediction. Specifically, DishFT-GNN trains a teacher model and a student model, iteratively. The teacher model learns to capture the correlation between distribution shifts of historical and future data, which is then utilized as intermediate supervision to guide the student model to learn future-aware spatiotemporal embeddings for accurate prediction. Through extensive experiments on two real-world datasets, we verify the state-of-the-art performance of DishFT-GNN.
A mechanical model of a laminated composite ring on a nonreciprocal elastic foundation is a valuable engineering tool during the early design stages of various applications, such as non-pneumatic wheels, flexible bearings, expandable tubulars in oil wells, and vascular stents interacting with blood vessel linings, especially under non-axisymmetric loadings. Despite its importance, limited research has focused on the interaction between laminated composite rings and nonreciprocal elastic foundations. Moreover, no quantitative studies have yet explored the influence of foundation stiffness on the ring deformation. This work aims to develop an analytical framework for a laminated composite ring supported by a nonreciprocal elastic foundation under non-axisymmetric loading conditions. The model generates a design map that correlates the foundation stiffness with the ring deformation, accounting for ring dimensions, laminate layup architecture, and lamina anisotropy. The closed-form solution provides an efficient design tool for analyzing non-axisymmetric and nonuniform loadings at a low computational cost. The resulting design map provides a valuable resource for exploring the interaction between the nonreciprocal foundation and the laminated ring. The proposed analytical framework and design map hold broad potential applications in automotive, mechanical, civil, and biomedical engineering fields.
A discrete Boltzmann model (DBM) for plasma kinetics is proposed. The constructing of DBM mainly considers two aspects. The first is to build a physical model with sufficient physical functions before simulation. The second is to present schemes for extracting more valuable information from massive data after simulation. For the first aspect, the model is equivalent to a magnetohydrodynamic model plus a coarse-grained model for the most relevant TNE behaviors including the entropy production rate. A number of typical benchmark problems including Orszag-Tang (OT) vortex problem are used to verify the physical functions of DBM. For the second aspect, the DBM use non-conserved kinetic moments of (f-feq) to describe non-equilibrium state and behaviours of complex system. The OT vortex problem and the Richtmyer-Meshkov instability (RMI) are practical applications of the second aspect. For RMI with interface inverse and re-shock process, it is found that, in the case without magnetic field, the non-organized momentum flux shows the most pronounced effects near shock front, while the non-organized energy flux shows the most pronounced behaviors near perturbed interface. The influence of magnetic field on TNE effects shows stages: before the interface inverse, the TNE strength is enhanced by reducing the interface inverse speed; while after the interface inverse, the TNE strength is significantly reduced. Both the global average TNE strength and entropy production rate contributed by non-organized energy flux can be used as physical criteria to identify whether or not the magnetic field is sufficient to prevent the interface inverse.
Direct laser acceleration of electrons in near-critical-density (NCD) carbon nanotube foams (CNFs) has its advantages in the high-efficiency generation of relativistic electrons and broadband X-rays. Here, we report the first simultaneous measurement on the spectra of laser-driven electrons and X-rays from CNFs at moderately relativistic intensities of around 5\times{10}^{19}\ W/cm^2.\ The density and thickness of the CNFs were scanned in the experiments, indicating the optimized electrons temperature of 5.5 MeV and X-ray critical energy of 5 keV. Two-dimensional (2D) particle-in-cell (PIC) simulations confirm that the electrons, with a temperature significantly higher than the pondermotive scale, are directly accelerated by the laser along the NCD plasma channel, while the bright X-rays are emitted by these electrons through betatron radiation or Thomson backscattering inside the channel. The simultaneously generated electrons and X-rays, automatically synchronized with the femtosecond laser driver, are suitable for applications such as bi-modal radiography.
Very thin free-flowing liquid sheets are promising targets for high-repetition-rate laser-ion acceleration. In this work, we report the generation of micrometer-thin free-flowing liquid sheets from the collision of two liquid jets, and study the vibration and jitter in their surface normal direction. The dependence of their motion amplitudes on the generation parameters is studied in detail. The origins of the vibration and jitter are discussed. Our results indicate that when the generation parameters are optimized, the motion amplitudes in the stable region can be stabilized below 3.7 μm to meet the stringent requirement of sheet position stability for a tight-focusing setup in laser-ion acceleration experiments.
Multimodal summarization (MS) aims to generate a summary from multimodal input. Previous works mainly focus on textual semantic coverage metrics such as ROUGE, which considers the visual content as supplemental data. Therefore, the summary is ineffective to cover the semantics of different modalities. This paper proposes a multi-task cross-modality learning framework (CISum) to improve multimodal semantic coverage by learning the cross-modality interaction in the multimodal article. To obtain the visual semantics, we translate images into visual descriptions based on the correlation with text content. Then, the visual description and text content are fused to generate the textual summary to capture the semantics of the multimodal content, and the most relevant image is selected as the visual summary. Furthermore, we design an automatic multimodal semantics coverage metric to evaluate the performance. Experimental results show that CISum outperforms baselines in multimodal semantics coverage metrics while maintaining the excellent performance of ROUGE and BLEU.
Mitosis nuclei count is one of the important indicators for the pathological diagnosis of breast cancer. The manual annotation needs experienced pathologists, which is very time-consuming and inefficient. With the development of deep learning methods, some models with good performance have emerged, but the generalization ability should be further strengthened. In this paper, we propose a two-stage mitosis segmentation and classification method, named SCMitosis. Firstly, the segmentation performance with a high recall rate is achieved by the proposed depthwise separable convolution residual block and channel-spatial attention gate. Then, a classification network is cascaded to further improve the detection performance of mitosis nuclei. The proposed model is verified on the ICPR 2012 dataset, and the highest F-score value of 0.8687 is obtained compared with the current state-of-the-art algorithms. In addition, the model also achieves good performance on GZMH dataset, which is prepared by our group and will be firstly released with the publication of this paper. The code will be available at: https://github.com/antifen/mitosis-nuclei-segmentation.
Post-acceleration of protons in helical coil targets driven by intense, ultrashort laser pulses can enhance the ion energy by utilizing the transient current originating from the self-discharging of the targets. The acceleration length of the protons can exceed a few millimeters, and the accelerating gradient is in the order of GeV/m. How to ensure the synchronization of the accelerating electric field with the protons is a crucial problem for an efficient post-acceleration. In this paper, we study how the electric field mismatch induced by the current dispersion affects the synchronous acceleration of the protons. We propose a scheme using a two-stage helical coil to control the current dispersion. With optimized parameters, the energy gain of protons is enhanced by 4 times. And it is expected that the proton energy would reach 45 MeV using a hundreds-terawatt laser, or over 100 MeV using a petawatt laser, by controlling the current dispersion.
Magnetic flux emergence from the solar interior to the atmosphere is believed to be a key process of formation of solar active regions and driving solar eruptions. Due to the limited capability of observation, the flux emergence process is commonly studied using numerical simulations. In this paper, we developed a numerical model to simulate the emergence of a twisted magnetic flux tube from the convection zone to the corona using the AMR--CESE--MHD code, which is based on the conservation-element solution-element method with adaptive mesh refinement. The result of our simulation agrees with that of many previous ones with similar initial conditions but using different numerical codes. In the early stage, the flux tube rises from the convection zone as driven by the magnetic buoyancy until it reaches close to the photosphere. The emergence is decelerated there and with piling-up of the magnetic flux, the magnetic buoyancy instability is triggered, which allows the magnetic field to partially enter into the atmosphere. Meanwhile, two gradually separated polarity concentration zones appear in the photospheric layer, transporting the magnetic field and energy into the atmosphere through their vortical and shearing motions. Correspondingly, the coronal magnetic field has also been reshaped to a sigmoid configuration containing a thin current layer, which resembles the typical pre-eruptive magnetic configuration of an active region. Such a numerical framework of magnetic flux emergence as established will be applied in future investigations of how solar eruptions are initiated in flux emergence active regions.
We consider the asymptotic behavior of the KPZ fixed point $\{\mathsf H(x,t)\}_{x\in\mathbb R, t>0}$ conditioned on $\mathsf H(0,T)=L$ as $L$ goes to infinity. The main result is a conditional limit theorem for the fluctuations of $\mathsf H$ in the region near the line segment connecting the origin $(0,0)$ and $(0,T)$ for both step and flat initial conditions. The limit random field can be represented as a functional of two independent Brownian bridges, and in addition the limit random field depends also on the initial law of the KPZ fixed point. In particular for temporal fluctuations, the limit process indexed by line segment between $(0,0)$ and $(0,T)$, when the KPZ is with step initial condition, has the law of the minimum of two independent Brownian bridges; and when the KPZ is with flat initial condition the limit process has the law of the minimum of two independent Brownian bridges, each in addition perturbed by a common Gaussian random variable. For spatial-temporal fluctuations, the conditional limit theorem sheds light on the asymptotic behaviors of the point-to-point geodesic of the directed landscape conditioned on its length and as the length tends to infinity.
Here we report the generation of MeV alpha-particles from H-11B fusion initiated by laser-accelerated boron ions. Boron ions with maximum energy of 6MeV and fluence of 10^9/MeV/sr@5MeV were generated from 60-nm-thick self-supporting boron nanofoils irradiated by 1J femtosecond pulses at an intensity of 10^19W/cm^2. By bombarding secondary hydrogenous targets with the boron ions, 3*10^5/sr alpha-particles from H-11B fusion were registered, which is consistent with the theoretical yield calculated from the measured boron energy spectra. Our results demonstrate an alternative way toward ultrashort MeV alpha-particle sources employing compact femtosecond lasers. The ion acceleration and product measurement scheme are referential for the studies on the ion stopping power and cross-section of the H-11B reaction in solid or plasma.