Robustness quantification method for network intrusion detection models

Deep learning techniques have been successfully applied to network intrusion detection tasks, but as in the case of autonomous driving and face recognition, the reliability of the system itself has become a pressing issue. Robustness is a key attribute to determine whether a deep learning system is secure and reliable, and we also choose to explore the security of intrusion detection models from a new perspective of robustness quantification. In this paper, we focus on the intrusion detection model based on long and short-term memory, and use a fine-grained linear approximation method to derive a more accurate robustness bound on the nonlinear activation function with tighter linear constraints. We can use this bound to quantitatively measure the robustness of the detection model and determine whether the model is susceptible to the influence of adversarial samples. In our experiments, we test networks with various structures on the MNIST dataset, and the results show that our proposed method can effectively deduce the robustness bounds of output elements, and has good scalability and applicability.     

Neutron skin thickness of 90Zr and symmetry energy constrained by charge exchange spin-dipole excitations

The charge exchange spin-dipole (SD) excitations of \begin{document}$ ^{90} $\end{document}Zr are studied using the Skyrme Hartee-Fock plus proton-neutron random phase approximation with SAMi-J interactions. The experimental value of the model-independent sum rule obtained from the SD strength distributions of \begin{document}$ ^{90} $\end{document}Zr(p, n)\begin{document}$ ^{90} $\end{document}Nb and \begin{document}$ ^{90} $\end{document}Zr(n, p)\begin{document}$ ^{90} $\end{document}Y is used to deduce the neutron skin thickness. The neutron skin thickness \begin{document}$ \Delta r_{np} $\end{document} of \begin{document}$ ^{90} $\end{document}Zr is extracted as \begin{document}$ 0.083\pm0.032 $\end{document} fm, which is similar to the results of other studies. Based on the correlation analysis of the neutron skin thickness \begin{document}$ \Delta r_{np} $\end{document} and the nuclear symmetry energy J as well as its slope parameter L, a constraint from the extracted \begin{document}$ \Delta r_{np} $\end{document} leads to the limitation of J to \begin{document}$ 29.2 \pm 2.6 $\end{document} MeV and L to \begin{document}$ 53.3 \pm 28.2 $\end{document} MeV.     

An emerging strategy for tracing the anti-erectile dysfunction components of raw and salt-processed semens of Astragali Complanati by combinatory liquid chromatography–mass spectrometry-based quantitative analysis,efficacy assessment on impotent rats,and partial least squares regression

Semens of Astragali Complanati own anti-erectile dysfunction effect; however, the components which contribute to the anti-erectile dysfunction effect remain unclear. This work raised a strategy that integrates liquid chromatography coupled mass spectrometry-based quantitative analysis, anti-erectile dysfunction assessment on impotent rats, and their relationship analysis for pinpointing anti-erectile dysfunction components from semens of Astragali Complanati. For simultaneous quantification of seven major components in raw and salt-processed semens of Astragali Complanati, an accurate and reliable liquid chromatography–mass spectrometry method was developed under multiple reaction monitoring mode. Of note, chloramphenicol was employed as the internal standard. The method showed good linearity and repeatability, where the recovery rates of each component ranged from 98.1 to 104.7%, and the precisions of intra- and interday were all within 3.4%. The method has been used for quantification of the seven major components in 10 batches of raw and salt-processed semens of Astragali Complanati. Then, the anti-erectile dysfunction effects of raw and salt-processed semens of Astragali Complanati were evaluated on impotent rats. Gray relationship analysis and partial least squares regression were combined for elucidating the relationship. As a result, complanatuside, astragalin, complanatoside B, and kaempferol were found to be responsible for anti-erectile dysfunction effect of Astragali Complanati.     

Discrimination of Beers by Cyclic Voltammetry Using a Single Carbon Screen-printed Electrode

A fast, simple and costless methodology without sample pre-treatment is proposed for the discrimination of beers. It is based on cyclic voltammetry (CV) using commercial carbon screen-printed electrodes (SPCE) and includes a correction of the signals measured with different SPCE units. Data are submitted to partial least squares discriminant analysis (PLS−DA) and support vector machine discriminant analysis (SVM−DA), which allow a reasonable classification of the beers. Also, CV data from beers can be used to predict their alcoholic degree by partial least squares (PLS) and artificial neural networks (ANN). In general, non-linear methods provide better results than linear ones.     

A parametric investigation of the friction performance of PC‐ABS parts processed by FDM additive manufacturing process

The friction performance is an important factor of parts processed by fused deposition modeling (FDM) for various engineering applications. It is one type of failure made of surface contact. The proper use of FDM process parameters can bring a significant reduction in friction and the amount of wear, thereby leading to a reduction in the material waste. To date, very little studies have been performed in this area. This paper investigates the effect of FDM manufacturing parameters on the friction performance of polycarbonate‐acrylonitrile butadiene styrene prototypes processed by FDM using definitive screening design and partial least squares method. The observation of surface morphology was obtained by the scanning electron microscopy to examine the effect of process parameters on the microstructure. The experimental results have shown that layer thickness, air gap, raster angle, and build orientation are the most influential factors affecting the friction performance of FDM manufactured parts. The proposed approach presented in this study provides an impetus to develop analytical modeling and functional relationships between FDM manufacturing parameters and friction performance. Copyright © 2017 John Wiley & Sons, Ltd.     

基于高通量解析算法的复杂样品重叠气相色谱-质谱信号的快速分析

化学计量学算法为重叠气相色谱-质谱(GC-MS)信号的解析提供了有效手段,但其在计算过程中一般需要将数据进行分段处理,然后只对信号的某些区间进行解析,难以实现真正意义上的高通量分析。该文结合移动窗口目标转换因子分析(MWTTFA)和非负免疫算法(NNIA),建立了一种高通量解析方法。首先,根据所有可能存在的目标组分的标准质谱信息,利用MWTTFA检验复杂信号中存在的组分,并确定目标组分的质谱信息和洗脱时间区域。以得到的质谱信息作为后续计算的输入值,利用NNIA解析得到相应的色谱信息。采用快速升温程序对17种和42种农药混合标准样品的GC-MS信号进行分析,利用所建立的方法可在10 min内得到全部组分的色谱和质谱信息。     

具有N-各向异性拉普拉斯算子的超线性问题的先验界

通过改进Brezis和Merle的方法,结合Moser-Trudinger不等式,移动平面方法及比较原理,得到了方程-Q_Nu=f(u),u∈W_0~(1,N)(Ω)的正解的先验界,其中Ω是R~N中的一个有界光滑区域,非线性项f至多具有指数型增长.