Complete mitochondrial genomes of two deep-sea pandalid shrimps,Heterocarpus ensifer and Bitias brevis:insights into the phylogenetic position of Pandalidae(Decapoda:Caridea)

The mitochondrial genome(mitogenome)analysis is a significant tool for investigating the evolutionary history of metazoan animals.The family Pandalidae is a diverse caridean group containing mainly deep-sea species.Until May 302019,only two complete mitogenomes are available in GenBank.Here we present the complete mitogenome sequences of two deep-sea pandalid shrimps,Heterocarpus ensifer and Bitias brevis through Illumina sequencing.The mitochondrial genomes were determined to be 15939 bp and 15891 bp long,and both consist of 13 protein-coding genes(PCGs),23 transfer-RNA genes(tRNAs),two ribosomal-RNA genes(rRNAs),and one control region.The nucleotide composition is biased toward adenine and thymine.Overall,the gene contents and arrangements are consistent with the pancrustacean ground pattern.The alignment of the control regions of four pandalids reveals a conserved sequence block(CSB)(104 bp in length,average GC%=29.47%and 69.23%similarity).A phylogenetic analysis based on 51 Caridea complete mitogenomes revealed that the deep-sea pandalid shrimps are situated an intermediate lineage,with a tendency to originated from those living in shallow sea area.

Analytical and Traveling Wave Solutions to the Fifth Order Standard Sawada-Kotera Equation via the Generalized exp(-Φ(ξ))-Expansion Method

In this article, we propose a generalized exp(-Φ(ξ))-expansion method and successfully implement it to find exact traveling wave solutions to the fifth order standard Sawada-Kotera (SK) equation. The exact traveling wave solutions are established in the form of trigonometric, hyperbolic, exponential and rational functions with some free parameters. It is shown that this method is standard, effective and easily applicable mathematical tool for solving nonlinear partial differential equations arises in the field of mathematical physics and engineering.

Decomposition Theorems for Semi-order Fuzzy Supermartingales and Submartingales

Based on semi - order fuzzy supermaringales andsubmartingales, the semi- order fuzzy supermartingaleand submartingale theory is developed. The main resultis to generalize the Doob decomposition and the Riesz de-composition theorems of standard martingale theory tosemi - order fuzzy supermaringales and submartingales.The structure of semi - order fuzzy supermaringales andsubmartingales and the conditions of that they has Doobdecomposition (resp. Riesz decomposition) are discussedin detail.

Deep Transfers of p-Class Tower Groups

Let p be a prime. For any finite p-group G, the deep transfers T H,G ' : H / H ' → G ' / G " from the maximal subgroups H of index (G:H) = p in G to the derived subgroup G ' are introduced as an innovative tool for identifying G uniquely by means of the family of kernels ùd(G) =(ker(T H,G ')) (G: H) = p. For all finite 3-groups G of coclass cc(G) = 1, the family ùd(G) is determined explicitly. The results are applied to the Galois groups G =Gal(F3 (∞)/ F) of the Hilbert 3-class towers of all real quadratic fields F = Q(√d) with fundamental discriminants d > 1, 3-class group Cl3(F) □ C3 × C3, and total 3-principalization in each of their four unramified cyclic cubic extensions E/F. A systematic statistical evaluation is given for the complete range 1 d 7, and a few exceptional cases are pointed out for 1 d 8.

O(t^(-β))-SYNCHRONIZATION AND ASYMPTOTIC SYNCHRONIZATION OF DELAYED FRACTIONAL ORDER NEURAL NETWORKS

This article explores the O(t^(-β))synchronization and asymptotic synchronization for fractional order BAM neural networks(FBAMNNs)with discrete delays,distributed delays and non-identical perturbations.By designing a state feedback control law and a new kind of fractional order Lyapunov functional,a new set of algebraic sufficient conditions are derived to guarantee the O(t^(-β))Synchronization and asymptotic synchronization of the considered FBAMNNs model;this can easily be evaluated without using a MATLAB LMI control toolbox.Finally,two numerical examples,along with the simulation results,illustrate the correctness and viability of the exhibited synchronization results.

On the Caginalp for a Conserve Phase-Field with a Polynomial Potentiel of Order 2<i>p</i>- 1

Our aim in this paper is to study on the Caginalp for a conserved phase-field with a polynomial potentiel of order 2p - 1. In this part, one treats the conservative version of the problem of generalized phase field. We consider a regular potential, more precisely a polynomial term of the order 2p - 1 with edge conditions of Dirichlet type. Existence and uniqueness are analyzed. More precisely, we precisely, we prove the existence and uniqueness of solutions.

Axisymmetric alternating direction explicit scheme for efficient coupled simulation of hydro-mechanical interaction in geotechnical engineering-Application to circular footing and deep tunnel in saturated ground

Explicit solution techniques have been widely used in geotechnical engineering for simulating the coupled hydro-mechanical(H-M) interaction of fluid flow and deformation induced by structures built above and under saturated ground, i.e. circular footing and deep tunnel. However, the technique is only conditionally stable and requires small time steps, portending its inefficiency for simulating large-scale H-M problems. To improve its efficiency, the unconditionally stable alternating direction explicit(ADE)scheme could be used to solve the flow problem. The standard ADE scheme, however, is only moderately accurate and is restricted to uniform grids and plane strain flow conditions. This paper aims to remove these drawbacks by developing a novel high-order ADE scheme capable of solving flow problems in nonuniform grids and under axisymmetric conditions. The new scheme is derived by performing a fourthorder finite difference(FD) approximation to the spatial derivatives of the axisymmetric fluid-diffusion equation in a non-uniform grid configuration. The implicit Crank-Nicolson technique is then applied to the resulting approximation, and the subsequent equation is split into two alternating direction sweeps,giving rise to a new axisymmetric ADE scheme. The pore pressure solutions from the new scheme are then sequentially coupled with an existing geomechanical simulator in the computer code fast Lagrangian analysis of continua(FLAC). This coupling procedure is called the sequentially-explicit coupling technique based on the fourth-order axisymmetric ADE scheme or SEA-4-AXI. Application of SEA-4-AXI for solving axisymmetric consolidation of a circular footing and of advancing tunnel in deep saturated ground shows that SEA-4-AXI reduces computer runtime up to 42%-50% that of FLAC’s basic scheme without numerical instability. In addition, it produces high numerical accuracy of the H-M solutions with average percentage difference of only 0.5%-1.8%.

Laplace NMR谱图重建——从经典正则化到深度学习

拉普拉斯核磁共振(Laplace NMR)可以提供待测样品的扩散系数或弛豫时间等物理参数信息,是用于研究分子化学结构、动力学和相互作用的强大工具.Laplace NMR的适用性很大程度上取决于拉普拉斯逆变换相关的信号处理算法的性能.在本文中,我们首先讨论了Laplace NMR谱图重建问题的不适定性,接着回顾了经典的基于正则化约束的重建算法,并介绍了目前前沿的深度学习算法在处理Laplace反演问题方面的应用,最后总结了这些算法的优缺点,并对Laplace NMR信号处理方法未来改进方向进行了展望.

High-Order BAM, Its Modifications and Performance Analyses

Based on Tai’ s high - order bidirectional associative memory ( HOBAM) and Stmposon’ s intraconnected BAM (IBAM) , two Improved models are first presented and discussed in this paper. The improved models not only retain the advantages of both HOBAM and ISAM but overcome the shortcomings of Kosko’ s BAM also. Secondly their recall stabilities in synchronous and asyn-chronous update modes have been proven by defining corresponding energy functions which decrease as the re-call process proceeds such that the systems can ensure all the training pattern pairs to become local minima of the energy surfaces. Finally with signal - to - noise ratio (SNR) approach, we show that their storage capacities and error correction capabilities are better than that of the HOBAM.

课堂深度学习的育人表征、实践价值与教学策略

深度学习以学生对知识的理解分析与迁移运用为价值取向,强调学生在真实问题情境中批判性学习,进而实现个体高阶思维和问题解决能力的发展。课堂深度学习具有促进学生高阶思维发展、注重个体行为的深层投入、聚焦学生情感的深度体验三个方面的表征。课堂深度学习有利于促进学生的有效发展、助推素养为本的课堂变革与支持学校育人的提质增效。因此,要树立导向深度学习的课堂理念,采用多元融合的教学方式,整合构建概念化与结构化的教学内容,以及创设与优化课堂教学情境。

基于深度强化学习的图书馆架序智能识别方法

为了解决现有图书馆架序识别方法存在的不足,设计基于深度强化学习的图书馆架序智能识别方法。利用深度强化学习架构,确定架序识别规则,计算图书馆架序参数,识别图书馆架序中书号字符数据,确定书号字符识别系数的取值范围,实现基于深度强化学习的图书馆架序智能识别。实验结果表明,应用文中提出的识别方法,可使图书资源排列架序与规定架序保持一致,识别准确度较高,有效解决了图书馆管理中的图书乱架问题,具有较好的应用性能。

基于深度卷积-递归神经网络的手绘草图识别方法

针对现有基于深度学习的手绘草图识别方法直接从整体上提取手绘草图的图像特征,而忽略了草图中笔画的顺序信息的问题,利用手绘草图的笔画顺序信息,将深度卷积神经网络与递归神经网络相结合,提出一种基于深度卷积-递归神经网络的手绘草图识别方法.首先按照绘画草图时的笔画顺序提取笔画,生成多幅子笔画草图,并形成一个笔画数依次递增的子笔画草图序列;然后采用深度卷积神经网络依次提取该序列中每一幅子笔画草图的图像特征,并将提取的图像特征按照原先子笔画草图排列的顺序进行排序,作为递归神经网络的输入;最后利用递归神经网络来构建不同图像特征间的时序关系,以提高手绘草图的识别准确率.在现有最大的手绘草图数据集TU-Berlin Sketch数据集上的实验结果表明,文中方法能有效地提升手绘草图的识别准确率.

学习科学视域下的e-Learning深度学习研究

深度学习是学习科学的主旨,也是现阶段所要破解的e-Learning难题,因此,对学习科学视域下的e-Learning深度学习进行研究,追溯起源,把握热点对推进e-Learning学习意义深远。在梳理e-Learning的现状及主要问题的基础之上,剖析了学习科学领域深度学习的重要性,以深度学习的内涵与特征、有意义学习是深度学习的主旨、高水平思维是深度学习的核心等理念为依据构建了e-Learning环境下深度学习分析模型,继而又以此为据,对e-Learning深度学习的研究现状进行述评。研究发现,为了进一步推动学习科学的发展,深度学习的研究必须在学习过程的评估与分析、情感体验等方面进一步加强,在研究方法与研究视角方面进行多角度创新,研究应该更精细深入,多进行相关的量化研究,以充分利用学习评估技术对深度学习的现状进行把脉。

基于深度张量投影网络的机械故障诊断方法研究

针对现有的、基于深度卷积神经网络的故障诊断方法利用池化层对高阶输入张量降维时容易破坏张量数据,造成数据信息丢失,以及网络结构相对复杂的不足,构造了一种深度张量投影网络。该网络利用张量投影层代替传统卷积神经网络中的池化层,在对输入的高阶张量数据进行降维时,不会对张量数据造成破坏,避免了特征信息的丢失,提高了模型对故障的识别准确率;并且张量投影层是一种维度可变的降维层,可以简化网络结构。在此基础上,结合高阶谱和深度张量投影网络各自的优点,提出了基于深度张量投影网络的机械故障诊断方法。在提出的方法中,利用高阶谱提取故障信号特征,将得到的高阶张量谱图输入到构建的深度张量投影网络模型中进行高阶张量降维和识别。提出的方法成功应用到齿轮箱故障诊断中。实验结果表明,所提方法能够更好地保留原始故障信息,有效识别不同类型的故障,准确率优于传统深度卷积神经网络故障诊断方法。

基于深度学习的快速长时视觉跟踪算法

在目标尺寸变化、遮挡和出视场等复杂长时视觉跟踪环境下,现有基于深度学习的视觉跟踪算法很难对目标进行实时准确的跟踪。针对该问题,提出一种快速的长时视觉跟踪算法,该算法由一个快速短时视觉跟踪算法和一个快速全局重检测模块组成。在基准算法SiamRPN中加入二阶通道与区域空间融合的注意力模块作为短时视觉跟踪算法,在保证快速性的同时,提高算法的短时视觉跟踪精确度和成功率;为使改进后的短时视觉跟踪算法具有快速的长时视觉跟踪能力,在算法中加入提出的基于模板匹配的全局重检测模块,该模块使用轻量级网络和快速的相似度判断方法,加快重检测速率。在OTB100、LaSOT、UAV20L、VOT2018-LT、VOT2020-LT等5个数据集上进行测试,实验结果表明,所提算法在长时视觉跟踪中具有优越的跟踪性能,平均速度达104帧/s。

全硅MCM-41担载的Ni-W催化剂上二苯并噻吩加氢脱硫反应动力学研究

以二苯并噻吩 (DBT)的质量分数为 0 .2 %～ 0 .8%的十氢萘溶液为模型化合物 ,在中压固定床反应器上 ,研究了全硅MCM 41担载的Ni W硫化物催化剂上的加氢脱硫反应 (HDS)动力学。理论分析和实验结果都表明 ,常用的Langmiur Hinshelwood方程不适用于气 液 固三相滴流床积分反应器。应用假 1级平推流反应模型 ,分别得到了各催化剂作用下的该反应的速率常数和表观活化能。催化剂的Ni/W摩尔比 (nNi/nW)为 0 .2 5、 0 .5、0 .75和 1.0时 ,该催化HDS反应的表观活化能分别为 117.3、 96.7、 64 .4和 72 .9kJ/mol。其中 ,nNi/nW =0 .75的反应表观活化能最低。将催化剂的HDS活性与表观活化能相关联可以发现 ,表观活化能和催化剂活性之间有对应关系 ,表观活化能越低 。

基于深度学习的水声通信信号调制类型识别

水声通信信号识别为水声通信侦察和对抗的重要前提,具有重要作用。然而,传统的水声通信信号识别方法通常是基于信号处理和模式识别技术,依赖领域专家的专业知识和经验进行特征选择和提取,具有较强的主观性,且可能无法利用更复杂的信号特征。本文基于深度学习提出一种水声通信信号识别的智能方法。首先利用仿真数据对卷积神经网络进行训练,然后分别使用仿真和湖上试验数据对算法网络进行测试。仿真结果表明,在SNR=5dB时,该方法对2ASK、4ASK、BPSK、QPSK、2FSK、4FSK和OFDM等7种水下通信信号的识别率均能达到90%以上,7种湖上试验的通信信号类型平均识别率达到97.9%。这表明该方法具有良好的宽容性。此外,本文还通过对基于高阶累积量和深度学习方法的比较,验证了本文提出方法具有显著的优越性。

深部金属矿山RS-TOPSIS岩爆预测模型及其应用

基于逼近理想解排序法的基本原理,选用岩石的单轴抗压强度σc、压拉比σc/σt、岩石的应力系数σθ/σc、弹性变形指数Wet 以及岩体的完整性系数Kv 等影响因素作为预测指标,以实际工程中的20个岩爆实例为样本,依据岩爆预测分级标准,运用粗糙集理论计算各预测指标支持度和权重,建立深部金属矿山的RS-TOPSIS岩爆预测模型,并在实际工程进行应用。结果显示,通过计算相对贴近度来确定分类区间的方法方便易行;建立RS-TOPSIS岩爆预测模型具有较高的准确度,预测精度达90%;对锌铜矿的岩爆预测结果与工程实际吻合,表明RS-TOPSIS岩爆预测方法具有工程实用价值,可以运用于深部金属矿山的岩爆预测。

我国e-Learning环境深度学习研究述评

建设学习型社会、构筑终身学习体系是国际教育大趋势,培养学习者在真实情景和多重技术环境下深度学习能力已成为重要的时代命题。本文对深度学习的产生和发展进行历史回朔,利用内容分析法从研究方向、研究范畴、研究类型及应用对象对国内10年来e-Learning环境下深度学习的相关刊文献进行梳理,同时对文献被引频次和关键词词频进行排序分析。研究得出4点结论:一是研究日趋多元与深入,呈现全面发展格局;二是实证研究得到重视,设计及行动研究有待深入;三是学习理论、技术观和教学理论与其存在密切关联;四是大视角、游戏与开放环境的研究成为未来关注重点。