• Stochastic microstructure reconstruction has become an indispensable part of computational materials science, but ongoing developments are specific to particular material systems. In this paper, we address this generality problem by presenting a transfer learning-based approach for microstructure reconstruction and structure-property predictions that is applicable to a wide range of material systems. The proposed approach incorporates an encoder-decoder process and feature-matching optimization using a deep convolutional network. For microstructure reconstruction, model pruning is implemented in order to study the correlation between the microstructural features and hierarchical layers within the deep convolutional network. Knowledge obtained in model pruning is then leveraged in the development of a structure-property predictive model to determine the network architecture and initialization conditions. The generality of the approach is demonstrated numerically for a wide range of material microstructures with geometrical characteristics of varying complexity. Unlike previous approaches that only apply to specific material systems or require a significant amount of prior knowledge in model selection and hyper-parameter tuning, the present approach provides an off-the-shelf solution to handle complex microstructures, and has the potential of expediting the discovery of new materials.
  • Interest in the superconducting proximity effect has recently been reignited by theoretical predictions that it could be used to achieve topological superconductivity. Low-T$_{c}$ superconductors have predominantly been used in this effort, but small energy scales of ~1 meV have hindered the characterization of the emergent electronic phase, limiting it to extremely low temperatures. In this work, we use molecular beam epitaxy to grow topological insulator Bi$_{2}$Te$_{3}$ in a range of thicknesses on top of a high-T$_{c}$ superconductor Fe(Te,Se). Using scanning tunneling microscopy and spectroscopy, we detect {\Delta}$_{ind}$ as high as ~3.5 meV, which is the largest reported gap induced by proximity to an s-wave superconductor to-date. We find that {\Delta}$_{ind}$ decays with Bi$_{2}$Te$_{3}$ thickness, but remains finite even after the topological surface states had been formed. Finally, by imaging the scattering and interference of surface state electrons, we provide a microscopic visualization of the fully gaped Bi$_{2}$Te$_{3}$ surface state due to Cooper pairing. Our results establish Fe-based high-T$_{c}$ superconductors as a promising new platform for realizing high-T$_{c}$ topological superconductivity.
  • Besides the text content, documents and their associated words usually come with rich sets of meta informa- tion, such as categories of documents and semantic/syntactic features of words, like those encoded in word embeddings. Incorporating such meta information directly into the generative process of topic models can improve modelling accuracy and topic quality, especially in the case where the word-occurrence information in the training data is insufficient. In this paper, we present a topic model, called MetaLDA, which is able to leverage either document or word meta information, or both of them jointly. With two data argumentation techniques, we can derive an efficient Gibbs sampling algorithm, which benefits from the fully local conjugacy of the model. Moreover, the algorithm is favoured by the sparsity of the meta information. Extensive experiments on several real world datasets demonstrate that our model achieves comparable or improved performance in terms of both perplexity and topic quality, particularly in handling sparse texts. In addition, compared with other models using meta information, our model runs significantly faster.
  • Understanding the effects of dust extinction is important to properly interpret observations. The optical total-to-selective extinction ratio, Rv = Av/E(B-V), is widely used to describe extinction variations in ultraviolet and optical bands. Since the Rv=3.1 extinction curve adequately represents the average extinction law of diffuse regions in the Milky Way, it is commonly used to correct observational measurements along sightlines toward diffuse regions in the interstellar medium. However, the Rv value may vary even along different diffuse interstellar medium sightlines. In this paper, we investigate the optical--mid-infrared (mid-IR) extinction law toward a very diffuse region at l = 165 in the Galactic plane, which was selected based on a CO emission map. Adopting red clump stars as extinction tracers, we determine the optical-to-mid-IR extinction law for our diffuse region in the two APASS bands (B, V), the three XSTPS-GAC bands (g, r, i), the three 2MASS bands (J, H, Ks, and the two WISE bands (W1, W2). Specifically, 18 red clump stars were selected from the APOGEE--RC catalog based on spectroscopic data in order to explore the diversity of the extinction law. We find that the optical extinction curves exhibit appreciable diversity. The corresponding Rv ranges from 1.7 to 3.8, while the mean Rv value of 2.8 is consistent with the widely adopted average value of 3.1 for Galactic diffuse clouds. There is no apparent correlation between Rv value and color excess E(B-V) in the range of interest, from 0.2 to 0.6 mag, or with specific visual extinction per kiloparsec, AV/d.
  • Relational data are usually highly incomplete in practice, which inspires us to leverage side information to improve the performance of community detection and link prediction. This paper presents a Bayesian probabilistic approach that incorporates various kinds of node attributes encoded in binary form in relational models with Poisson likelihood. Our method works flexibly with both directed and undirected relational networks. The inference can be done by efficient Gibbs sampling which leverages sparsity of both networks and node attributes. Extensive experiments show that our models achieve the state-of-the-art link prediction results, especially with highly incomplete relational data.
  • This paper aims at synthesizing filamentary structured images such as retinal fundus images and neuronal images, as follows: Given a ground-truth, to generate multiple realistic looking phantoms. A ground-truth could be a binary segmentation map containing the filamentary structured morphology, while the synthesized output image is of the same size as the ground-truth and has similar visual appearance to what have been presented in the training set. Our approach is inspired by the recent progresses in generative adversarial nets (GANs) as well as image style transfer. In particular, it is dedicated to our problem context with the following properties: Rather than large-scale dataset, it works well in the presence of as few as 10 training examples, which is common in medical image analysis; It is capable of synthesizing diverse images from the same ground-truth; Last and importantly, the synthetic images produced by our approach are demonstrated to be useful in boosting image analysis performance. Empirical examination over various benchmarks of fundus and neuronal images demonstrate the advantages of the proposed approach.
  • Intrinsic colors of normal stars are derived in the popularly used infrared bands involving the 2MASS/JHKs, WISE, Spitzer/IRAC and AKARI/S9W filters. Based on three spectroscopic surveys -- LAMOST, RAVE and APOGEE, stars are classified into groups of giants and dwarfs, as well as metal-normal and metal-poor stars. An empirical analytical relation of the intrinsic color is obtained with stellar effective temperature (Teff) for each group of stars after the zero-reddening stars are selected from the blue edge in the $J-\lambda$ versus (Teff) diagram. It is found that metallicity has little effect on the infrared colors. In the near-infrared bands, our results agree with previous work. In addition, the color indexes H-W2 and Ks-W1 that are taken as constant to calculate interstellar extinction are discussed. The intrinsic color of M-type stars are derived separately due to lack of accurate measurement of their effective temperature.
  • A new accelerator complex, HIAF (the High Intensity Heavy Ion Accelerator Facility), has been approved in China. It is designed to provide intense primary and radioactive ion beams for research in high energy density physics, nuclear physics, atomic physics as well as other applications. In order to achieve a high intensity of up to 5e11 ppp 238U34+, the Compression Ring (CRing) needs to stack more than 5 bunches transferred from the Booster Ring (BRing). However, the normal bucket to bucket injection scheme can only achieve an intensity gain of 2, so an injection method, fixed barrier bucket (BB) supported by electron cooling, is proposed. To suppress the severe space charge effect during the stacking process, off-alignment is adopted in the cooler to control the transverse emittance. In this paper, simulation and optimization with the BETACOOL program are presented.
  • A long-standing issue in topological insulator research has been to find a material that provides an ideal platform for characterizing topological surface states without interference from bulk electronic states and can reliably be fabricated as bulk crystals. This material would be a bulk insulator, have a surface state Dirac point energy well isolated from the bulk valence and conduction bands, have high surface state electronic mobility, and be growable as large, high quality bulk single crystals. Here we show that this major materials obstacle in the field is overcome by crystals of lightly Sn-doped Bi1.1Sb0.9Te2S (Sn-BSTS) grown by the Vertical Bridgeman method, which we characterize here via angle-resolved photoemission spectroscopy, scanning tunneling microscopy, transport studies of the bulk and surface states, and X-ray diffraction and Raman scattering. We present this new material as a bulk topological insulator that can be reliably grown and studied in many laboratories around the world.
  • Several solenoids are usually installed in electron cooler device to guide the motion of the electron beam in the cooler. However, the solenoids also have influence to the ion beam in the cooler storage ring. The transverse motion of the ion beam in storage ring will become coupled, if the solenoids installed in the electron cooler are not compensated perfectly. In this paper, the coupled transverse motion due to the uncompensated cooler's solenoids of CSRm (The main storage ring in the IMP, Lan Zhou, China) is studied, and the coupled beam envelopes are calculated by a new method.
  • Electron cooler is used to improve the quality of the beam in synchrotron, however it also introduces nonlinear electromagnetic field, which cause tuneshift, tunespread and may drive resonances leading to beam loss. In this paper the tuneshift and the tunespread caused by nonlinear electromagnetic field of the hollow electron beam was investigated, and the resonance driving terms of the nonlinear electromagnetic field was analysed. The differences were presented comparing with the solid electron beam. The calculations were performed for $^{238}U^{32+}$ ions of energy 1.272MeV stored in CSRm, using the parameters given in table1. The conclusion is that in this situation nonlinear field caused by the hollow electron beam do not lead to serious resonances.
  • The correction of closed orbit has great influence on the operation of synchrotron. The design of correction system is one significant component of lattice design. It is suggested to set BPMs at the peaks of betatron oscillation. The correctors need to be located at the positions where \b{eta} function is large or the sources of large errors. In the simulation of the closed orbit correction of HIMM synchrotron, one important reason affecting the result of horizontal correction is the longitudinal alignment error of dipole magnet. It is advisable to decrease this kind of alignment error while the deflection angle of dipole magnet is large.