Search

In financial field, a robust software system is of vital importance to ensure the smooth operation of financial transactions. However, many financial corporations still depend on operators to identify and eliminate the system failures when financial software systems break down. This traditional operation method is time consuming and extremely inefficient. To improve the efficiency and accuracy of system failure detection and thereby reduce the impact of system failures on financial services, we propose a novel machine learning-based framework to predict the occurrence of system exceptions and failures in a financial software system. In particular, we first extract rich information from system logs and eliminate noises in the data. Then the cleaned data is leveraged as the input of our proposed anomaly prediction framework which consists of three modules: key performance indicator(KPI) data prediction module, anomaly identification module and severity classification module. Notably, we design a hierarchical architecture of alarm classifiers and try to alleviate the influence of class-imbalance problem on the overall performance. Empirically, the experimental results demonstrate the superior performance of our proposed method on a real-world financial software system log data set.

Unsupervised cross-spectral stereo matching aims at recovering disparity given cross-spectral image pairs without any supervision in the form of ground truth disparity or depth. The estimated depth provides additional information complementary to individual semantic features, which can be helpful for other vision tasks such as tracking, recognition and detection. However, there are large appearance variations between images from different spectral bands, which is a challenge for cross-spectral stereo matching. Existing deep unsupervised stereo matching methods are sensitive to the appearance variations and do not perform well on cross-spectral data. We propose a novel unsupervised cross-spectral stereo matching framework based on image-to-image translation. First, a style adaptation network transforms images across different spectral bands by cycle consistency and adversarial learning, during which appearance variations are minimized. Then, a stereo matching network is trained with image pairs from the same spectra using view reconstruction loss. At last, the estimated disparity is utilized to supervise the spectral-translation network in an end-to-end way. Moreover, a novel style adaptation network F-cycleGAN is proposed to improve the robustness of spectral translation. Our method can tackle appearance variations and enhance the robustness of unsupervised cross-spectral stereo matching. Experimental results show that our method achieves good performance without using depth supervision or explicit semantic information., Comment: accepted by AAAI-19

With the rapid development of urbanization, the boom of vehicle numbers has resulted in serious traffic accidents, which led to casualties and huge economic losses. The ability to predict the risk of traffic accident is important in the prevention of the occurrence of accidents and to reduce the damages caused by accidents in a proactive way. However, traffic accident risk prediction with high spatiotemporal resolution is difficult, mainly due to the complex traffic environment, human behavior, and lack of real-time traffic-related data. In this study, we collected big traffic accident data. By analyzing the spatial and temporal patterns of traffic accident frequency, we presented the spatiotemporal correlation of traffic accidents. Based on the patterns we found in analysis, we proposed a high accurate deep learning model based on recurrent neural network toward the prediction of traffic accident risk. The predictive accident risk can be potential applied to the traffic accident warning system. The proposed method can be integrated into an intelligent traffic control system toward a more reasonable traffic prediction and command organization.

The recent discovery of a Weyl semimetal in TaAs offers the first Weyl fermion observed in nature and dramatically broadens the classification of topological phases. However, in TaAs it has proven challenging to study the rich transport phenomena arising from emergent Weyl fermions. The series Mo$_x$W$_{1-x}$Te$_2$ are inversion-breaking, layered, tunable semimetals already under study as a promising platform for new electronics and recently proposed to host Type II, or strongly Lorentz-violating, Weyl fermions. Here we report the discovery of a Weyl semimetal in Mo$_x$W$_{1-x}$Te$_2$ at $x = 25\%$. We use pump-probe angle-resolved photoemission spectroscopy (pump-probe ARPES) to directly observe a topological Fermi arc above the Fermi level, demonstrating a Weyl semimetal. The excellent agreement with calculation suggests that Mo$_x$W$_{1-x}$Te$_2$ is the first Type II Weyl semimetal. We also find that certain Weyl points are at the Fermi level, making Mo$_x$W$_{1-x}$Te$_2$ a promising platform for transport and optics experiments on Weyl semimetals.

It has recently been proposed that electronic band structures in crystals give rise to a previously overlooked type of Weyl fermion, which violates Lorentz invariance and, consequently, is forbidden in particle physics. It was further predicted that Mo$_x$W$_{1-x}$Te$_2$ may realize such a Type II Weyl fermion. One crucial challenge is that the Weyl points in Mo$_x$W$_{1-x}$Te$_2$ are predicted to lie above the Fermi level. Here, by studying a simple model for a Type II Weyl cone, we clarify the importance of accessing the unoccupied band structure to demonstrate that Mo$_x$W$_{1-x}$Te$_2$ is a Weyl semimetal. Then, we use pump-probe angle-resolved photoemission spectroscopy (pump-probe ARPES) to directly observe the unoccupied band structure of Mo$_x$W$_{1-x}$Te$_2$. For the first time, we directly access states $> 0.2$ eV above the Fermi level. By comparing our results with $\textit{ab initio}$ calculations, we conclude that we directly observe the surface state containing the topological Fermi arc. Our work opens the way to studying the unoccupied band structure as well as the time-domain relaxation dynamics of Mo$_x$W$_{1-x}$Te$_2$ and related transition metal dichalcogenides., Comment: Incorporates earlier results presented in arXiv:1512.09099, by the same authors

Here we repair the single-layer MoSe2 field-effect transistors by the EDTA processing, after which the devices' room-temperature carrier mobility increases from 0.1 to over 70cm2/Vs. The atomic dynamics is constructed by the combined study of the first-principle calculation, aberration-corrected transmission electron microscopy and Raman spectroscopy. Single/double Se vacancies are revealed originally, which cause some mid-gap impurity states and localize the device carriers. They are found repaired with the result of improved electronic transport. Such a picture is confirmed by a 1.5cm-1 red shift in the Raman spectra.

Weyl semimetals have sparked intense research interest, but experimental work has been limited to the TaAs family of compounds. Recently, a number of theoretical works have predicted that compounds in the Mo$_x$W$_{1-x}$Te$_2$ series are Weyl semimetals. Such proposals are particularly exciting because Mo$_x$W$_{1-x}$Te$_2$ has a quasi two-dimensional crystal structure well-suited to many transport experiments, while WTe$_2$ and MoTe$_2$ have already been the subject of numerous proposals for device applications. However, with available ARPES techniques it is challenging to demonstrate a Weyl semimetal in Mo$_x$W$_{1-x}$Te$_2$. According to the predictions, the Weyl points are above the Fermi level, the system approaches two critical points as a function of doping, there are many irrelevant bulk bands, the Fermi arcs are nearly degenerate with bulk bands and the bulk band gap is small. Here, we study Mo$_x$W$_{1-x}$Te$_2$ for $x = 0.07$ and 0.45 using pump-probe ARPES. The system exhibits a dramatic response to the pump laser and we successfully access states $> 0.2$eV above the Fermi level. For the first time, we observe direct, experimental signatures of Fermi arcs in Mo$_x$W$_{1-x}$Te$_2$, which agree well with theoretical calculations of the surface states. However, we caution that the interpretation of these features depends sensitively on free parameters in the surface state calculation. We comment on the prospect of conclusively demonstrating a Weyl semimetal in Mo$_x$W$_{1-x}$Te$_2$., Comment: 24 pages, 4 figures