Search

We explore a scenario of interacting dynamical dark energy model with the interaction term $Q$ including the varying equation-of-state parameter $w$. Using the data combination of the cosmic microwave background, the baryon acoustic oscillation, and the type Ia supernovae, to global fit the interacting dynamical dark energy model, we find that adding a factor of the varying $w$ in the function of $Q$ can change correlations between the coupling constant $\beta$ and other parameters, and then has a huge impact on the fitting result of $\beta$. In this model, the fitting value of $H_{0}$ is lower at the $3.54 \sigma$ level than the direct measurement value of $H_{0}$ . Comparing to the case of interacting dynamical dark energy model with $Q$ excluding $w$, the model with $Q$ including the constant $w$ is more favored by the current mainstream observation. To obtain higher fitting values of $H_{0}$ and narrow the discrepancy of $H_{0}$ between different observations, additional parameters including the effective number of relativistic species, the total neutrino mass, and massive sterile neutrinos are considered in the interacting dynamical dark energy cosmology. We find that the $H_{0}$ tension can be further reduced in these models, but is still at the about $3 \sigma$ level., Comment: 14 pages, 4 figures

We study three-body loss in an ultracold mixture of a thermal Bose gas and a degenerate Fermi gas. We find that at unitarity, where the interspecies scattering length diverges, the usual inverse-square temperature scaling of the three-body loss found in non-degenerate systems is strongly modified and reduced with the increasing degeneracy of the Fermi gas. While the reduction of loss is qualitatively explained within the few-body scattering framework, a remaining suppression provides evidence for the long-range RKKY interactions mediated by fermions between bosons. Our model based on RKKY interactions quantitatively reproduces the data without free parameters, and predicts one order of magnitude reduction of the three-body loss coefficient in the deeply Fermi-degenerate regime.

With the miniaturization and integration of nanoelectronic devices, efficient heat removal becomes a key factor affecting the reliable operation of the nanoelectronic device. With the high intrinsic thermal conductivity, good mechanical flexibility, and precisely controlled growth, two-dimensional (2D) materials are widely accepted as ideal candidates for thermal management materials. In this work, by solving the phonon Boltzmann transport equation (BTE) based on first-principles calculations, we comprehensively investigated the thermal conductivity of novel 2D layered MSi$_2$N$_4$ (M = Mo, W). Our results point to competitive thermal conductivities (162 W/mK) of monolayer MoSi$_2$N$_4$, which is around two times larger than that of WSi$_2$N$_4$ and seven times larger than that of silicene despite their similar non-planar structures. It is revealed that the high thermal conductivity arises mainly from its large group velocity and low anharmonicity. Our result suggests that MoSi$_2$N$_4$ could be a potential candidate for 2D thermal management materials.

Deep learning has an increasing impact to assist research, allowing, for example, the discovery of novel materials. Until now, however, these artificial intelligence techniques have fallen short of discovering the full differential equation of an experimental physical system. Here we show that a dynamical neural network, trained on a minimal amount of data, can predict the behavior of spintronic devices with high accuracy and an extremely efficient simulation time, compared to the micromagnetic simulations that are usually employed to model them. For this purpose, we re-frame the formalism of Neural Ordinary Differential Equations (ODEs) to the constraints of spintronics: few measured outputs, multiple inputs and internal parameters. We demonstrate with Spin-Neural ODEs an acceleration factor over 200 compared to micromagnetic simulations for a complex problem -- the simulation of a reservoir computer made of magnetic skyrmions (20 minutes compared to three days). In a second realization, we show that we can predict the noisy response of experimental spintronic nano-oscillators to varying inputs after training Spin-Neural ODEs on five milliseconds of their measured response to different excitations. Spin-Neural ODE is a disruptive tool for developing spintronic applications in complement to micromagnetic simulations, which are time-consuming and cannot fit experiments when noise or imperfections are present. Spin-Neural ODE can also be generalized to other electronic devices involving dynamics., Comment: 16 pages, 4 figures

AI-based methods have been widely applied to tourism demand forecasting. However, current AI-based methods are short of the ability to process long-term dependency, and most of them lack interpretability. The Transformer used initially for machine translation shows an incredible ability to long-term dependency processing. Based on the Transformer, we proposed a time series Transformer (Tsformer) with Encoder-Decoder architecture for tourism demand forecasting. The proposed Tsformer encodes long-term dependency with encoder, captures short-term dependency with decoder, and simplifies the attention interactions under the premise of highlighting dominant attention through a series of attention masking mechanisms. These improvements make the multi-head attention mechanism process the input sequence according to the time relationship, contributing to better interpretability. What's more, the context processing ability of the Encoder-Decoder architecture allows adopting the calendar of days to be forecasted to enhance the forecasting performance. Experiments conducted on the Jiuzhaigou valley and Siguniang mountain tourism demand datasets with other nine baseline methods indicate that the proposed Tsformer outperformed all baseline models in the short-term and long-term tourism demand forecasting tasks. Moreover, ablation studies demonstrate that the adoption of the calendar of days to be forecasted contributes to the forecasting performance of the proposed Tsformer. For better interpretability, the attention weight matrix visualization is performed. It indicates that the Tsformer concentrates on seasonal features and days close to days to be forecast in short-term forecasting.

Split learning is a promising privacy-preserving distributed learning scheme that has low computation requirement at the edge device but has the disadvantage of high communication overhead between edge device and server. To reduce the communication overhead, this paper proposes a loss-based asynchronous training scheme that updates the client-side model less frequently and only sends/receives activations/gradients in selected epochs. To further reduce the communication overhead, the activations/gradients are quantized using 8-bit floating point prior to transmission. An added benefit of the proposed communication reduction method is that the computations at the client side are reduced due to reduction in the number of client model updates. Furthermore, the privacy of the proposed communication reduction based split learning method is almost the same as traditional split learning. Simulation results on VGG11, VGG13 and ResNet18 models on CIFAR-10 show that the communication cost is reduced by 1.64x-106.7x and the computations in the client are reduced by 2.86x-32.1x when the accuracy degradation is less than 0.5% for the single-client case. For 5 and 10-client cases, the communication cost reduction is 11.9x and 11.3x on VGG11 for 0.5% loss in accuracy., Comment: Accepted by SIPS '21

Workflow decision making is critical to performing many practical workflow applications. Scheduling in edge-cloud environments can address the high complexity problem of workflow applications, while decreasing the data transmission delay between the cloud and end devices. However, because of the heterogeneous resources in edge-cloud environments and the complicated data dependencies among the tasks in a workflow, significant challenges for workflow scheduling remain, including the selection of an optimal tasks-servers solution from the possible numerous combinations. The existing studies have been mainly done subject to rigorous conditions without fluctuations, ignoring the fact that workflow scheduling is typically present in uncertain environments. In this study, we focus on reducing the execution cost of workflow applications mainly caused by task computation and data transmission, while satisfying the workflow deadline in uncertain edge-cloud environments. The Triangular Fuzzy Numbers (TFNs) are adopted to represent the task processing time and data transferring time. A cost-driven fuzzy scheduling strategy based on an Adaptive Discrete Particle Swarm Optimization (ADPSO) algorithm is proposed, which employs the operators of Genetic Algorithm (GA). This strategy introduces the randomly two-point crossover operator, neighborhood mutation operator, and adaptive multipoint mutation operator of GA to effectively avoid converging on local optima. The experimental results show that our strategy can effectively reduce the workflow execution cost in uncertain edge-cloud environments, compared with other benchmark solutions.

We demonstrate the transfer of $^{23}$Na$^{40}$K molecules from a closed-channel dominated Feshbach-molecule state to the absolute ground state. The Feshbach molecules are initially created from a gas of sodium and potassium atoms via adiabatic ramping over a Feshbach resonance at 78.3$\,$G. The molecules are then transferred to the absolute ground state using stimulated Raman adiabatic passage with an intermediate state in the spin-orbit-coupled complex $|c^3 \Sigma^+, v=35, J=1 \rangle \sim |B^1\Pi, v=12, J=1\rangle$. Our measurements show that the pump transition dipole moment linearly increases with the closed-channel fraction. Thus, the pump-beam intensity can be two orders of magnitude lower than is necessary with open-channel dominated Feshbach molecules. We also demonstrate that the phase noise of the Raman lasers can be reduced by filter cavities, significantly improving the transfer efficiency., Comment: 8 pages, 7 figures

Small-scale magnetic fields are not only the fundamental element of the solar magnetism, but also closely related to the structure of the solar atmosphere. The observations have shown that there is a ubiquitous tangled small-scale magnetic field with a strength of 60 $\sim$ 130\,G in the canopy forming layer of the quiet solar photosphere. On the other hand, the multi-dimensional MHD simulations show that the convective overshooting expels the magnetic field to form the magnetic canopies at a height of about 500\,km in the upper photosphere. However, the distribution of such small-scale ``canopies" in the solar photosphere cannot be rigorously constrained by either observations and numerical simulations. Based on stellar standard models, we identify that these magnetic canopies can act as a global magnetic-arch splicing layer, and find that the reflections of the solar p-mode oscillations at this magnetic-arch splicing layer results in significant improvement on the discrepancy between the observed and calculated p-mode frequencies. The location of the magnetic-arch splicing layer is determined at a height of about 630\,km, and the inferred strength of the magnetic field is about 90\,G. These features of the magnetic-arch splicing layer derived independently in the present study are quantitatively in agreement with the presence of small-scale magnetic canopies as those obtained by the observations and 3-D MHD simulations., Comment: 16 pages, 5 figures, accepted for publication in ApJ

This paper presents a fundamental analysis connecting phase noise and long-term frequency accuracy of oscillators and explores the possibilities and limitations in crystal-less frequency calibration for wireless edge nodes from a noise-impact perspective. N-period-average jitter (NPAJ) is introduced as a link between the spectral characterization of phase noise and long-term frequency accuracy. It is found that flicker noise or other colored noise profiles coming from the reference in a frequency synthesizer is the dominant noise source affecting long-term frequency accuracy. An average processing unit embedded in an ADPLL is proposed based on the N-period-average jitter concept to enhance frequency accuracy in a Calibrate and Open-loop scenario commonly used in low power radios. With this low-cost block, the frequency calibration accuracy can be directly associated with the reference noise performance. Thus, the feasibility of XO-less design with certain communication standards can be easily evaluated with the proposed theory., Comment: 6 pages, 8 figures