基于深度张量神经网络预测有机分子的相互作用能（英文）

分子的相互作用在分子动力学模拟过程中起着关键的作用.受限于计算资源,大分子的长时间尺度的相互作用能无法通过量化计算实现.本文采用一种深度学习框架-深度张量神经网络来预测三个有机分子相关体系中量化精度的相互作用能.其中,分子的几何结构和原子类型作为网络的输入用于预测相互作用能.通过分层生成的数据集合实现了网络中隐层参数的优化和训练.相互作用能的预测结果显示,深度张量神经网络可以在较短的时间内,在1 kcal/mol的平均绝对误差的范围内准确预测分子间的相互作用能.这一过程提高了计算效率,并为计算相互作用能提供了可靠的计算框架.     

基于TorchMD的粗粒化分子动力模拟研究

粗粒化模型通过简化原子性质以及原子间的相互作用实现生物大分子长时间尺度的分子动力学模拟. 深度学习通过模拟人类的认知过程实现海量数据的准确分类和回归过程. 本论文将这两种技术进行融合,利用基于深度学习的粗粒化分子动力学模拟技术研究分子在不同状态之间的变化过程,并提出基于TorchMD的分子动力学模拟的分析框架. 在本工作中,MFDP聚类算法被用于在三维的CV变量空间中进行聚类,并确定分子的若干主要状态,在完成聚类的同时,给出各类中的代表分子构象,并给出类之间的分子构象. 这为后续利用String算法分析分子在不同状态间的转换路径打下基础. 通过String算法,迭代搜索得到分子在不同状态之间的变化路径以及对应的势能变化曲线. 通过与已有文献的结果进行对比,验证了基于TorchMD的粗粒化分子动力学模拟的理论框架可以在相对较短的时间尺度里研究分子的变化过程.     

张量网络与神经网络在物理学中的应用和交融

基于张量网络的数值重正化群方法,被广泛地应用到物理学的研究中,已经成为量子多体计算方法大家庭的重要一员。近年来,基于神经网络的机器学习方法也逐渐渗透到物理学领域,并被成功应用在量子多体等问题的研究中。文章简要综述了近年来张量网络和神经网络在凝聚态物理和统计物理学的应用,并讨论了两者的相互交叉和结合。     

类锂离子里德堡态中的自旋-其它轨道相互作用

以多电子原子精细结构哈密顿的球张量形式和三价原子的非相对论性能级结构理论为基础,借助不可约张量理论,建立了计算类锂离子里德堡态中的自旋-其它轨道相互作用能的一种解析理论形式.完成了所有的角向积分和自旋求和计算,自旋-其它轨道相互作用能最终用径向积分来表示.应用所建立的理论对类锂离子(1s2np)2Pj态的自旋-其它轨道相互作用能进行了具体分析.     

类铍离子1s~22snp~3P(n=2～6,Z=5～8)态能级和精细结构参数的理论计算

在准相对论框架下,以多电子原子精细结构哈密顿的球张量形式为基础,借助不可约张量和角动量耦合理论,导出类铍离子1s22snp(n=2-6)3P态精细结构(包括自旋-轨道相互作用、自旋-其它轨道相互作用和自旋-自旋相互作用)和精细结构参数的解析表达式,并利用我们所开发的程序,对各项进行了具体地计算,计算结果与文献符合地较好.     

神经网络超参数优化的删除垃圾神经元策略

随着深度学习处理问题的日益复杂,神经网络的层数、神经元个数、和神经元之间的连接逐渐增加,参数规模急剧膨胀,优化超参数来提高神经网络的预测性能成为一个重要的任务.文献中寻找最优参数的方法如灵敏度剪枝、网格搜索等,算法复杂而且计算量庞大.本文提出一种超参数优化的“删除垃圾神经元策略”.权重矩阵中权重均值小的神经元,在预测中的贡献可以忽略,称为垃圾神经元.该策略就是通过删除这些垃圾神经元得到精简的网络结构,来有效缩短计算时间,同时提高预测准确率和模型泛化能力.采用这一策略,长短期记忆网络模型对几种典型混沌动力系统的预测性能得到显著改善.     

有限温度量子多体系统与热态张量网络

量子多体系统热力学性质的精确模拟在理论和实验方面都具有重要的价值。局域相互作用量子多体系统的热态满足互信息(mutual information)面积律,对于这样的系统,热态张量网络可以提供满足面积律的精确"波函数"拟设,提供了模拟有限温度系统的有力手段。文章介绍了关联格点模型在有限温度下的热态张量网络刻画及相关模拟方法。作者按照世界线热态张量网络和级数展开热态张量网络来分别介绍,并讨论了自由能极小变分原理与重正化群剪裁的优化原则。世界线框架内,人们发展了转移矩阵重正化群,基于纯化策略的有限温度密度矩阵重正化群,以及张量网络的线性重正化群等方法。在此基础上,介绍作者新近提出的级数展开热态张量网络方法,该方法受随机级数展开量子蒙特卡罗方法的启发,突破了世界线方法的局限,提高了有限温度计算重正化群模拟的精度标准,并且在计算阻挫量子自旋链模型时不会有负符号问题。此外,文章讨论了在两维格点系统上推广有限温度张量网络计算的进展和未来展望。     

氦原子2 3P态自旋-其他轨道相互作用的理论计算

借助不可约张量理论,导出了氦原子2^3P态自旋-其他轨道相互作用能的理论计算式,在推导过程中完成了所有角向积分和自旋求和计算．     

Ne-CH4分子间相互作用势的局域密度近似计算

利用基于密度泛函理论框架下的局域密度近似方法对Ne-CH4分子间的相互作用势进行了计算. 发现 当Ne原子和CH4分子之间的距离约为5.8 a.u.时, 计算的势能曲线存在最小值, 对应的势阱深度约为0.053 eV. 计算结果与实验值符合较好.     

Ne-CH4分子间相互作用势的局域密度近似计算

利用基于密度泛函理论框架下的局域密度近似方法对Ne-CH4分子间的相互作用势进行了计算. 发现: 当Ne原子和CH4分子之间的距离约为5.8 a.u.时, 计算的势能曲线存在最小值, 对应的势阱深度约为0.053 eV. 计算结果与实验值符合较好.     

Fast prediction of aerodynamic noise induced by the flow around a cylinder based on deep neural network

Accurate and fast prediction of aerodynamic noise has always been a research hotspot in fluid mechanics and aeroacoustics. The conventional prediction methods based on numerical simulation often demand huge computational resources, which are difficult to balance between accuracy and efficiency. Here, we present a data-driven deep neural network (DNN) method to realize fast aerodynamic noise prediction while maintaining accuracy. The proposed deep learning method can predict the spatial distributions of aerodynamic noise information under different working conditions. Based on the large eddy simulation turbulence model and the Ffowcs Williams-Hawkings acoustic analogy theory, a dataset composed of 1216 samples is established. With reference to the deep learning method, a DNN framework is proposed to map the relationship between spatial coordinates, inlet velocity and overall sound pressure level. The root-mean-square-errors of prediction are below 0.82 dB in the test dataset, and the directivity of aerodynamic noise predicted by the DNN framework are basically consistent with the numerical simulation. This work paves a novel way for fast prediction of aerodynamic noise with high accuracy and has application potential in acoustic field prediction.     

Deep-learning-based prediction of living cells mitosis via quantitative phase microscopy

We present a deep learning approach for living cells mitosis classification based on label-free quantitative phase imaging with transport of intensity equation methods. In the approach, we applied a pretrained deep convolutional neural network using transfer learning for binary classification of mitosis and non-mitosis. As a validation, we demonstrated the performances of the network trained by phase images and intensity images, respectively. The convolutional neural network trained by phase images achieved an average accuracy of 98.9% on the validation data, which outperforms the average accuracy 89.6% obtained by the network trained by intensity images. We believe that the quantitative phase microscopy in combination with deep learning enables researchers to predict the mitotic status of living cells noninvasively and efficiently.     

基于深度学习的两分量BEC中量子相变点的识别

识别物质的相变是物理学研究中一个重要问题.本文采用了一种混淆标签方案的卷积神经网络算法来识别两分量玻色-爱因斯坦凝聚(BEC)中量子相变点,通过计算神经网络输出的准确率,得到W型性能曲线,此性能曲线中间的极大值对应着量子相变的临界点.研究结果表明,深度学习得到的量子相变点与解析计算值吻合度较高.此混淆标签方案的深度学习研究方法可以应用到存在两种相的相变体系.     

Solutions to the Dirac equation for symmetric and asymmetric trigonometric Rosen-Morse potential using SUSYQM

In this paper, we calculate the energy spectra and the corresponding wavefunction for the symmetric and asymmetric trigonometric Rosen-Morse potential of the Dirac equation within the framework of spin and pseudospin symmetry limits including the tensor interaction using the supersymmetric quantum mechanics (SUSYQM) formalism. We have also reported some numerical results and figures to show the effect of tensor interaction.     

深度学习在核物理中的应用

深度学习是目前最好的模式识别工具,预期会在核物理领域帮助科学家从大量复杂数据中寻找与某些物理最相关的特征。本文综述了深度学习技术的分类,不同数据结构对应的最优神经网络架构,黑盒模型的可解释性与预测结果的不确定性。介绍了深度学习在核物质状态方程、核结构、原子核质量、衰变与裂变方面的应用,并展示如何训练神经网络预测原子核质量。结果发现使用实验数据训练的神经网络模型对未参与训练的实验数据拥有良好的预测能力。基于已有的实验数据外推,神经网络对丰中子的轻原子核质量预测结果与宏观微观液滴模型有较大偏离。此区域可能存在未被宏观微观液滴模型包含的新物理,需要进一步的实验数据验证。     

Measuring Dynamics in Evacuation Behaviour with Deep Learning

Bounded rationality is one crucial component in human behaviours. It plays a key role in the typical collective behaviour of evacuation, in which heterogeneous information can lead to deviations from optimal choices. In this study, we propose a framework of deep learning to extract a key dynamical parameter that drives crowd evacuation behaviour in a cellular automaton (CA) model. On simulation data sets of a replica dynamic CA model, trained deep convolution neural networks (CNNs) can accurately predict dynamics from multiple frames of images. The dynamical parameter could be regarded as a factor describing the optimality of path-choosing decisions in evacuation behaviour. In addition, it should be noted that the performance of this method is robust to incomplete images, in which the information loss caused by cutting images does not hinder the feasibility of the method. Moreover, this framework provides us with a platform to quantitatively measure the optimal strategy in evacuation, and this approach can be extended to other well-designed crowd behaviour experiments.     

Vibrational spectrum at a water surface: a hybrid quantum mechanics/molecular mechanics molecular dynamics approach

A hybrid quantum mechanics/molecular mechanics (QM/MM) molecular dynamics (MD) simulation is applied to the calculation of surface orientational structure and vibrational spectrum (second-order nonlinear susceptibility) at the vapor/water interface for the first time. The surface orientational structure of the QM water molecules is consistent with the previous MD studies, and the calculated susceptibility reproduces the experimentally reported one, supporting the previous results using the classical force field MD simulation. The present QM/MM MD simulation also demonstrates that the positive sign of the imaginary part of the second-order nonlinear susceptibility at the lower hydrogen bonding OH frequency region originates not from individual molecular orientational structure, but from cooperative electronic structure through the hydrogen bonding network.     

碰撞诱导离解(CID)(Ⅰ)——含时量子理论

本文采用含时量子力学微扰理论研究了分子碰撞诱导离解(CID)动力学过程。分子系统具有Morse振子和分子间指数排斥相互作用势的类型。所建立的分析理论公式能方便地用来讨论有关CID过程的各种物理化学问题。以(I₂,He)分子系统为例所进行的计算表明,在总能量一定的条件下,分子振动能比相对平动能更有效地促进CID过程的进行。而且,除必须克服离解能量势垒外,CID过程并没有额外的动力学阈值限制。 关键词：     

Local tensor network for strongly correlated projective States

The success of tensor network approaches in simulating strongly correlated quantum systems crucially depends on whether the many-body states that are relevant for the problem can be encoded in a local tensor network. Despite numerous efforts, strongly correlated projective states, including fractional quantum Hall states, have not yet found a local tensor network representation. Here we show that one can encode the calculation of averages of local operators in a Grassmann tensor network which is local. Our construction is explicit and allows the use of physically motivated trial wave functions as starting points in tensor network variational calculations.