The study of infall motion helps us to understand the initial stages of star formation. In this paper, we use the IRAM 30-m telescope to make mapping observations of 24 infall sources confirmed in previous work. The lines we use to track gas infall m
otions are HCO+ (1-0) and H13CO+ (1-0). All 24 sources show HCO+ emissions, while 18 sources show H13CO+ emissions. The HCO+ integrated intensity maps of 17 sources show clear clumpy structures; for the H13CO+ line, 15 sources show clumpy structures. We estimated the column density of HCO+ and H13CO+ using the RADEX radiation transfer code, and the obtained [HCO+]/[H2] and [H13CO+]/[HCO+] of these sources are about 10^-11 ~ 10^-7 and 10^-3~1, respectively. Based on the asymmetry of the line profile of the HCO+, we distinguish these sources: 19 sources show blue asymmetric profiles, and the other sources show red profiles or symmetric peak profiles. For eight sources that have double-peaked blue line profiles and signal-to-noise ratios greater than 10, the RATRAN model is used to fit their HCO^+ (1-0) lines, and to estimate their infall parameters. The mean Vin of these sources are 0.3 ~ 1.3 km/s, and the Min are about 10^-3 ~ 10^-4 Msun/yr , which are consistent with the results of intermediate or massive star formation in previous studies. The Vin estimated from the Myers model are 0.1 ~ 1.6 km/s, and the Min are within 10^-3 ~ 10^-5 Msun/yr. In addition, some identified infall sources show other star-forming activities, such as outflows and maser emissions. Especially for those sources with a double-peaked blue asymmetric profile, most of them have both infall and outflow evidence.
In this paper, by modeling the point cloud registration task as a Markov decision process, we propose an end-to-end deep model embedded with the cross-entropy method (CEM) for unsupervised 3D registration. Our model consists of a sampling network mod
ule and a differentiable CEM module. In our sampling network module, given a pair of point clouds, the sampling network learns a prior sampling distribution over the transformation space. The learned sampling distribution can be used as a good initialization of the differentiable CEM module. In our differentiable CEM module, we first propose a maximum consensus criterion based alignment metric as the reward function for the point cloud registration task. Based on the reward function, for each state, we then construct a fused score function to evaluate the sampled transformations, where we weight the current and future rewards of the transformations. Particularly, the future rewards of the sampled transforms are obtained by performing the iterative closest point (ICP) algorithm on the transformed state. By selecting the top-k transformations with the highest scores, we iteratively update the sampling distribution. Furthermore, in order to make the CEM differentiable, we use the sparsemax function to replace the hard top-$k$ selection. Finally, we formulate a Geman-McClure estimator based loss to train our end-to-end registration model. Extensive experimental results demonstrate the good registration performance of our method on benchmark datasets.
This paper proposes a novel matrix rank-one decomposition for quaternion Hermitian matrices, which admits a stronger property than the previous results in (sturm2003cones,huang2007complex,ai2011new). The enhanced property can be used to drive some im
proved results in joint numerical range, $mathcal{S}$-Procedure and quadratically constrained quadratic programming (QCQP) in the quaternion domain, demonstrating the capability of our new decomposition technique.
Single photon sources with high brightness and subnanosecond lifetimes are key components for quantum technologies. Optical nanoantennas can enhance the emission properties of single quantum emitters, but this approach requires accurate nanoscale pos
itioning of the source at the plasmonic hotspot. Here, we use plasmonic nanoantennas to simultaneously trap single colloidal quantum dots and enhance their photoluminescence. The nano-optical trapping automatically locates the quantum emitter at the nanoantenna hotspot without further processing. Our dedicated nanoantenna design achieves a high trap stiffness of 0.6 fN/nm/mW for quantum dot trapping, together with a relatively low trapping power of 2 mW/$mu$m$^2$. The emission from the nanoantenna-trapped single quantum dot shows 7x increased brightness, 50x reduced blinking, 2x shortened lifetime and a clear antibunching below 0.5 demonstrating true single photon emission. Combining nano-optical tweezers with plasmonic enhancement is a promising route for quantum technologies and spectroscopy of single nano-objects.
Entity extraction is a key technology for obtaining information from massive texts in natural language processing. The further interaction between them does not meet the standards of human reading comprehension, thus limiting the understanding of the
model, and also the omission or misjudgment of the answer (ie the target entity) due to the reasoning question. An effective MRC-based entity extraction model-MRC-I2DP, which uses the proposed gated attention-attracting mechanism to adjust the restoration of each part of the text pair, creating problems and thinking for multi-level interactive attention calculations to increase the target entity It also uses the proposed 2D probability coding module, TALU function and mask mechanism to strengthen the detection of all possible targets of the target, thereby improving the probability and accuracy of prediction. Experiments have proved that MRC-I2DP represents an overall state-of-the-art model in 7 from the scientific and public domains, achieving a performance improvement of up to compared to the model model in F1.
Point cloud registration is a fundamental problem in 3D computer vision. In this paper, we cast point cloud registration into a planning problem in reinforcement learning, which can seek the transformation between the source and target point clouds t
hrough trial and error. By modeling the point cloud registration process as a Markov decision process (MDP), we develop a latent dynamic model of point clouds, consisting of a transformation network and evaluation network. The transformation network aims to predict the new transformed feature of the point cloud after performing a rigid transformation (i.e., action) on it while the evaluation network aims to predict the alignment precision between the transformed source point cloud and target point cloud as the reward signal. Once the dynamic model of the point cloud is trained, we employ the cross-entropy method (CEM) to iteratively update the planning policy by maximizing the rewards in the point cloud registration process. Thus, the optimal policy, i.e., the transformation between the source and target point clouds, can be obtained via gradually narrowing the search space of the transformation. Experimental results on ModelNet40 and 7Scene benchmark datasets demonstrate that our method can yield good registration performance in an unsupervised manner.
Microlaser with multiple lasing bands is critical in various applications, such as full-colour display, optical communications and computing. Here, we propose a simple and efficient method for homogeneously doping rare earth elements into a silica wh
ispering-gallery-mode microcavity. By this method, we demonstrate simultaneous and stable lasing covering ultraviolet, visible and near-infrared bands in an ultrahigh-Q (exceeding 108) Er-Yb co-doped silica microsphere under room temperature and continuous-wave pump for the first time. The lasing thresholds of the 380, 410, 450, 560, 660, 800, 1080 and 1550 nm-bands are estimated to be 380, 150, 2.5, 12, 0.17, 1.7, 10 and 38 {mu}W, respectively, where the lasing in the 380, 410 and 450 nm-bands by Er element have not been separately demonstrated under room temperature and continuous-wave pump until this work. This ultrahigh-Q doped microcavity is an excellent platform for high-performance multi-band microlasers, ultrahigh-precise sensors, optical memories and cavity-enhanced light-matter interaction studies.
In this paper, we introduce a proximal-proximal majorization-minimization (PPMM) algorithm for nonconvex tuning-free robust regression problems. The basic idea is to apply the proximal majorization-minimization algorithm to solve the nonconvex proble
m with the inner subproblems solved by a sparse semismooth Newton (SSN) method based proximal point algorithm (PPA). We must emphasize that the main difficulty in the design of the algorithm lies in how to overcome the singular difficulty of the inner subproblem. Furthermore, we also prove that the PPMM algorithm converges to a d-stationary point. Due to the Kurdyka-Lojasiewicz (KL) property of the problem, we present the convergence rate of the PPMM algorithm. Numerical experiments demonstrate that our proposed algorithm outperforms the existing state-of-the-art algorithms.
Plasmonic nano-optical tweezers enable the non-invasive manipulation of nano-objects under low illumination intensities, and have become a powerful tool for nanotechnology and biophysics. However, measuring the trap stiffness of nanotweezers remains
a complicated task, which hinders the development of plasmonic trapping. Here, we describe an experimental method to measure the trap stiffness based on the temporal correlation of the fluorescence from the trapped object. The method is applied to characterize the trap stiffness in different double nanohole apertures and explore the influence of their design parameters in relationship with numerical simulations. Optimizing the double nanohole design achieves a trap stiffness 10x larger than the previous state-of-the-art. The experimental method and the design guidelines discussed here offer a simple and efficient way to improve the performance of nano-optical tweezers.
Plasmonic nano-tweezers use intense electric field gradients to generate optical forces able to trap nano-objects in liquids. However, part of the incident light is absorbed into the metal, and a supplementary thermophoretic force acting on the nano-
object arises from the resulting temperature gradient. Plasmonic nano-tweezers thus face the challenge of disentangling the intricate contributions of the optical and thermophoretic forces. Here, we show that commonly added surfactants can unexpectedly impact the trap performance by acting on the thermophilic or thermophobic response of the nano-object. Using different surfactants in double nanohole plasmonic trapping experiments, we measure and compare the contributions of the thermophoretic and the optical forces, evidencing a trap stiffness 20x higher using sodium dodecyl sulfate (SDS) as compared to Triton X-100. This work uncovers an important mechanism in plasmonic nano-tweezers and provides guidelines to control and optimize the trap performance for different plasmonic designs.
