IM-UFF: Extending the universal force field for interactive molecular modeling

The universal force field (UFF) is a broadly applicable classical force field that contains parameters for almost every atom type of the periodic table. This force field is non-reactive, i.e. the topology of the system under study is considered as fixed and no creation or breaking of covalent bonds is possible. This paper introduces interactive modeling-UFF (IM-UFF), an extension of UFF that combines the possibility to significantly modify molecular structures (as with reactive force fields) with a broad diversity of supported systems thanks to the universality of UFF. Such an extension lets the user easily build and edit molecular systems interactively while being guided by physics based inter-atomic forces. This approach introduces weighted atom types and weighted bonds, used to update topologies and atom parameterizations at every time step of a simulation. IM-UFF has been evaluated on a large set of benchmarks and is proposed as a self-contained implementation integrated in a new module for the SAMSON software platform for computational nanoscience available at http://www.samson-connect.net.     

基于CRFs和歧义模型的越南语分词

通过对越南语词法特点的研究,把越南语的基本特征融入到条件随机场中(Condition random fields,CRFs),提出了一种基于CRFs和歧义模型的越南语分词方法。通过机器标注、人工校对的方式获取了25 981条越南语分词语料作为CRFs的训练语料。越南语中交叉歧义广泛分布在句子中,为了克服交叉歧义的影响,通过词典的正向和逆向匹配算法从训练语料中抽取了5 377条歧义片段,并通过最大熵模型训练得到一个歧义模型,并融入到分词模型中。把训练语料均分为10份做交叉验证实验,分词准确率达到了96.55%。与已有越南语分词工具VnTokenizer比较,实验结果表明该方法提高了越南语分词的准确率、召回率和F值。     

结合深度学习与条件随机场的遥感图像分类

目的 为进一步提高遥感影像的分类精度,将卷积神经网络（CNN）与条件随机场（CRF）两个模型结合,提出一种新的分类方法。方法 首先采用CNN对遥感图像进行预分类,并将其类成员概率定义为CRF模型的一阶势函数;然后利用高斯核函数的线性组合定义CRF模型的二阶势函数,用全连接的邻域结构代替常见的4邻域或8邻域;接着加入区域约束,使用Mean-shift分割方法得到超像素,通过计算超像素的后验概率均值修正各像素的分类结果,鼓励连通区域结果的一致性;最后采用平均场近似算法实现整个模型的推断。结果 选用3组高分辨率遥感图像进行地物分类实验。本文方法不仅能抑制更多的分类噪声,同时还可以改善过平滑现象,保护各类地物的边缘信息。实验采用类精度、总体分类精度OA、平均分类精度AA,以及Kappa系数4个指标进行定量分析,与支持向量机（SVM）、CNN和全连接CRF相比,最终获得的各项精度均得到显著提升,其中,AA提高3.28个百分点,OA提高3.22个百分点,Kappa提高5.07个百分点。结论 将CNN与CRF两种模型融合,不仅可以获得像元本质化的特征,而且同时还考虑了图像的空间上下文信息,使分类更加准确,后加入的约束条件还能进一步保留地物目标的局部信息。本文方法适用于遥感图像分类领域,是一种精确有效的分类方法。     

机器人自然语言导航的层叠式条件随机场模型

提出了一种基于层叠条件随机场进行救灾机器人自然语言导航命令理解的方法。该方法由三层条件随机场(CRFs)构成:第一层用于导航词性标注,选取词、词性以及上下文作为特征模板生成导航词性标签;第二层用于导航过程提取,选择词、导航词性标签以及上下文构建特征模板生成导航过程标签;第三层用于起点终点识别,选取词、导航词性标签、导航过程标签以及上下文构建特征模板判断出地名词为起点还是终点。根据导航词性与导航要素的对应关系便可从命令中提取出导航信息。该方法能够处理完全不受限的自然语言导航命令,总体正确率达到78.6%,无需依赖特定的指令与地图,对完成救灾机器人导航的人机交互任务具有重要意义。     

A modern workflow for force-field development – Bridging quantum mechanics and atomistic computational models

In this article we present our recent efforts in designing a comprehensive consistent scientific workflow, nicknamed Wolf₂ Pack, for force-field optimization in the field of computational chemistry. Atomistic force fields represent a multiscale bridge that connects high-resolution quantum mechanics knowledge to coarser molecular mechanics-based models. Force-field optimization has so far been a time-consuming and error-prone process, and is a topic where the use of a scientific workflow can provide obvious great benefits. As a case study we generate a gas-phase force field for methanol using Wolf₂ Pack, with special attention given toward deriving partial atomic charges.     

A note on Gao’s algorithm for polynomial factorization

Shuhong Gao (2003) [6] has proposed an efficient algorithm to factor a bivariate polynomial f over a field F. This algorithm is based on a simple partial differential equation and depends on a crucial fact: the dimension of the polynomial solution space G associated with this differential equation is equal to the number r of absolutely irreducible factors of f. However, this holds only when the characteristic of F is either zero or sufficiently large in terms of the degree of f. In this paper we characterize a vector subspace of G for which the dimension is r, regardless of the characteristic of F, and the properties of Gao’s construction hold. Moreover, we identify a second vector subspace of G that leads to an analogous theory for the rational factorization of f.     

基于互补模型的汉语韵律间断自动检测

自动韵律间断检测和标注对语音理解和语音合成有十分重要的作用。提出了利用声学、词典和语法相关特征的互补模型方法检测汉语韵律间断。该方法具有下列优点:(1)摒弃了声学相关特征和词典、语法相关特征的独立性假设;(2)互补模型方法不仅在特征层上利用当前音节的上下文信息,而且在模型层次上利用了当前音节的上下文信息。在ASCCD语料库上验证了该方法能够获得90.34%的韵律间断的检测准确率,较基线系统有 6.09%的提高。     

基于多LCS和人工势场法的机器人行为控制

基于遗传算法的早熟收敛、局部最优解、占据较大的存储空间等缺陷,提出了一种基于多LCS和人工势场法的机器人行为控制方法,设计了特殊的适应度函数。不同算法的仿真实验结果表明多LCS和人工势场法相结合用于机器人行为控制是收敛的,很大程度上改善了遗传算法的早熟收敛和收敛速度慢等问题。     

An enhanced error estimator on the constitutive relation for plasticity problems

This paper presents an extension to plasticity problems of a new error estimator on the constitutive relation which was introduced by P. Ladevèze in a recent paper. Numerical experiments show that this enhanced error estimator can lead to a significant improvement in the effectivity indexes in the case of anisotropic meshes, or when perfect plasticity is approached.     

The use of genetic algorithms for assessment of high voltage fields

There are several numerical methods for calculation of electric fields but they need some sort of experience and trial and error to get the correct solution. A genetic algorithm (GA) based approach is used to eliminate the need for the experience and to save time and effort spent in the trial and error. Two enhancement techniques, namely, Refreshment Method and No-Twins Method, are used with the GA operators to improve the performance of the GA in assessment of high voltage fields. The performance of these two enhancement techniques has been studied for computing the electrostatic field in high voltage applications.