An Overview and Perspectives On Bidirectional Intelligence: Lmser Duality, Double IA Harmony,and Causal Computation

Advances on bidirectional intelligence are overviewed along three threads,with extensions and new perspectives.The first thread is about bidirectional learning architecture,exploring five dualities that enable Lmser six cognitive functions and provide new perspectives on which a lot of extensions and particularlly flexible Lmser are proposed.Interestingly,either or two of these dualities actually takes an important role in recent models such as U-net,ResNet,and Dense Net.The second thread is about bidirectional learning principles unified by best yIng-yAng(IA)harmony in BYY system.After getting insights on deep bidirectional learning from a bird-viewing on existing typical learning principles from one or both of the inward and outward directions,maximum likelihood,variational principle,and several other learning principles are summarised as exemplars of the BYY learning,with new perspectives on advanced topics.The third thread further proceeds to deep bidirectional intelligence,driven by long term dynamics(LTD)for parameter learning and short term dynamics(STD)for image thinking and rational thinking in harmony.Image thinking deals with information flow of continuously valued arrays and especially image sequence,as if thinking was displayed in the real world,exemplified by the flow from inward encoding/cognition to outward reconstruction/transformation performed in Lmser learning and BYY learning.In contrast,rational thinking handles symbolic strings or discretely valued vectors,performing uncertainty reasoning and problem solving.In particular,a general thesis is proposed for bidirectional intelligence,featured by BYY intelligence potential theory(BYY-IPT)and nine essential dualities in architecture,fundamentals,and implementation,respectively.Then,problems of combinatorial solving and uncertainty reasoning are investigated from this BYY IPT perspective.First,variants and extensions are suggested for AlphaGoZero like searching tasks,such as traveling salesman problem(TSP)and attributed graph matching(AGM)that are turned into Go like problems with help of a feature enrichment technique.Second,reasoning activities are summarized under guidance of BYY IPT from the aspects of constraint satisfaction,uncertainty propagation,and path or tree searching.Particularly,causal potential theory is proposed for discovering causal direction,with two roads developed for its implementation.

基于Transformer的多尺度遥感语义分割网络

为了提升遥感图像语义分割效果,本文针对分割目标类间方差小、类内方差大的特点,从全局上下文信息和多尺度语义特征2个关键点提出一种基于Transformer的多尺度遥感语义分割网络(muliti-scale Transformer network,MSTNet)。其由编码器和解码器2个部分组成,编码器包含基于Transformer改进的视觉注意网络(visual attention network,VAN)主干和基于空洞空间金字塔池化(atrous spatial pyramid pooling, ASPP)结构改进的多尺度语义特征提取模块(multi-scale semantic feature extraction module, MSFEM)。解码器采用轻量级多层感知器(multi-layer perception,MLP)配合编码器设计,充分分析所提取的包含全局上下文信息和多尺度表示的语义特征。MSTNet在2个高分辨率遥感语义分割数据集ISPRS Potsdam和LoveDA上进行验证,平均交并比(mIoU)分别达到79.50%和54.12%,平均F1-score(m F1)分别达到87.46%和69.34%,实验结果验证了本文所提方法有效提升了遥感图像语义分割的效果。

Transforming growth factor-β and toll-like receptor-4 polymorphisms are not associated with fibrosis in haemochromatosis

AIM:To investigate the role of genetic polymorphisms in the progression of hepatic fibrosis in hereditary haemochromatosis.METHODS:A cohort of 245 well-characterised C282Y homozygous patients with haemochromatosis was studied,with all subjects having liver biopsy data and DNA available for testing.This study assessed the association of eight single nucleotide polymorphisms(SNPs)in a total of six genes including toll-like receptor 4(TLR4),transforming growth factor-beta(TGF-β),oxoguanine DNA glycosylase,monocyte chemoattractant protein 1,chemokine C-C motif receptor 2 and interleukin-10 with liver disease severity.Genotyping was performed using high resolution melt analysis and sequencing.The results were analysed in relation to the stage of hepatic fibrosis in multivariate analysis incorporating other cofactors including alcohol consumption and hepatic iron concentration.RESULTS:There were significant associations between the cofactors of male gender(P=0.0001),increasing age(P=0.006),alcohol consumption(P=0.0001),steatosis(P=0.03),hepatic iron concentration(P<0.0001)and the presence of hepatic fibrosis.Of the candidate gene polymorphisms studied,none showed a significant association with hepatic fibrosis in univariate or multivariate analysis incorporating cofactors.We also specifically studied patients with hepatic iron loading above threshold levels for cirrhosis and compared the genetic polymorphisms between those with no fibrosis vs cirrhosis however there was no significant effect from any of the candidate genes studied.Importantly,in this large,well characterised cohort of patients there was no association between SNPs for TGF-βor TLR4and the presence of fibrosis,cirrhosis or increasing fibrosis stage in multivariate analysis.CONCLUSION:In our large,well characterised group of haemochromatosis subjects we did not demonstrate any relationship between candidate gene polymorphisms and hepatic fibrosis or cirrhosis.

Oxidation Kinetics of Aluminum Powders in a Gas Fluidized Bed Reactor in the Potential Application of Surge Arresting Materials

In this technical paper, the oxidation mechanism and kinetics of aluminum powders are discussed in great details. The potential applications of spherical aluminum powders after oxidation to be part of the surging arresting materials are discussed. Theoretical calculations of oxidation of spherical aluminum powders in a typical gas fluidization bed are demonstrated. Computer software written by the author is used to carry out the basic calculations of important parameters of a gas fluidization bed at different temperatures. A mathematical model of the dynamic system in a gas fluidization bed is developed and the analytical solution is obtained. The mathematical model can be used to estimate aluminum oxide thickness at a defined temperature. The mathematical model created in this study is evaluated and confirmed consistently with the experimental results on a gas fluidization bed. Detail technical discussion of the oxidation mechanism of aluminum is carried out. The mathematical deviations of the mathematical modeling have demonstrated in great details. This mathematical model developed in this study and validated with experimental results can bring a great value for the quantitative analysis of a gas fluidization bed in general from a theoretical point of view. It can be applied for the oxidation not only for aluminum spherical powders, but also for other spherical metal powders. The mathematical model developed can further enhance the applications of gas fluidization technology. In addition to the development of mathematical modeling of a gas fluidization bed reactor, the formation of oxide film through diffusion on both planar and spherical aluminum surfaces is analyzed through a thorough mathematical deviation using diffusion theory and Laplace transformation. The dominant defects and their impact to oxidation of aluminum are also discussed in detail. The well-controlled oxidation film on spherical metal powders such as aluminum and other metal spherical powders can potentially become an important part of switch devices of surge arresting materials, in general.

The assessment of time dependent flow of Williamson fluid with radiative blood flow against a wedge

The present pagination reports both Brownian diffusion and thermophoresis aspects subject to magneto hydrodynamic Williamson fluid model.Assuming the flow is unsteady and blood is treated as Williamson fluid over a wedge with radiation.The governing equations are transformed into ordinary differential equations by using similarity variables.The analytical solutions of the transformed governing equations are obtained by using the RK 4th order method along with shooting technique solver.The effects of various physical parameters such as Hartmann number,local Weissenberg number,radiation parameter,unsteadiness parameter,Prandtl number,Lewis number,Brownian diffusion,thermophoresis,wedge angle parameter,moving wedge parameter,on velocity,temperature,concentration,skin friction,heat transfer rate and mass transfer rate have been discussed in detail.The velocity and temperature profile deprives for larger We and an opposite trend is observed for concentration.The radiation parameter is propositional to temperature and a counter behaviour is observed for Pr.

融合CNN与Transformer的MRI脑肿瘤图像分割

为解决卷积神经网络(convolutional neural network,CNN)在学习全局上下文信息和边缘细节方面受到很大限制的问题,提出一种同时学习局语义信息和局部空间细节的级联神经网络用于脑肿瘤医学图像分割。首先将输入体素分别送入CNN和Transformer分支,在编码阶段结束后,采用一种双分支融合模块将2个分支学习到的特征有效地结合起来以实现全局信息与局部信息的融合。双分支融合模块利用哈达玛积对双分支特征之间的细粒度交互进行建模,同时使用多重注意力机制充分提取特征图通道和空间信息并抑制无效的噪声信息。在BraTS竞赛官网评估了本文方法,在BraTS2019验证集上增强型肿瘤区、全肿瘤区和肿瘤核心区的Dice分数分别为77.92%,89.20%和81.20%。相较于其他先进的三维医学图像分割方法,本文方法表现出了更好的分割性能,为临床医生做出准确的脑肿瘤细胞评估和治疗方案提供了可靠依据。

Large Power Transformer Fault Diagnosis and Prognostic Based on DBNC and D-S Evidence Theory

Power transformer is a core equipment of power system, which undertakes the important functions of power transmission and transformation, and its safe and stable operation has great significance to the normal operation of the whole power system. Due to the complex structure of the transformer, the use of single information for condition-based maintenance (CBM) has certain limitations, with the help of advanced sensor monitoring and information fusion technology, multi-source information is applied to the prognostic and health management (PHM) of power transformer, which is an important way to realize the CBM of power transformer. This paper presents a method which combine deep belief network classifier (DBNC) and D-S evidence theory, and it is applied to the PHM of the large power transformer. The experimental results show that the proposed method has a high correct rate of fault diagnosis for the power transformer with a large number of multi-source data.

融合Transformer和交互注意力网络的方面级情感分类模型

现有的大多数研究者使用循环神经网络与注意力机制相结合的方法进行方面级情感分类任务。然而,循环神经网络不能并行计算,并且模型在训练过程中会出现截断的反向传播、梯度消失和梯度爆炸等问题,传统的注意力机制可能会给句子中重要情感词分配较低的注意力权重。针对上述问题,该文提出了一种融合Transformer和交互注意力网络的方面级情感分类模型。首先利用BERT(bidirectional encoder representation from Transformers)预训练模型来构造词嵌入向量,然后使用Transformer编码器对输入的句子进行并行编码,接着使用上下文动态掩码和上下文动态权重机制来关注与特定方面词有重要语义关系的局部上下文信息。最后在5个英文数据集和4个中文评论数据集上的实验结果表明,该文所提模型在准确率和F1上均表现最优。

Transformer Characteristics of Linear Motor-Transformer Apparatus

The characteristics of linear transformer are studied analytically. The transformer is composed in one of modes of linear motor-transformer apparatus proposed for future wireless light rail vehicle (LRV). The secondary (onboard) power factor can be adjusted at any value by an onboard converter. The equivalent circuit is used to study the transferred power control. The parameters are determined by three-dimensional finite element method (FEM) analysis for one pole-pair model. Under the rated primary (input) and secondary voltage and current, which are specified for linear motor operation, the characteristics of the secondary power factor are cleared. It is also shown that the input capacitor can improve the primary power factor and decrease the input power capacity, but does not change the efficiency. This linear transformer has the efficiency of 91% and the input power factor of 0.87 when the apparatus without input capacitor is controlled at the secondary power factor of 0.4.

In-situ observations on interphase boundary migration and grain growth during α/γ phase transformation in iron-4.2%Cr alloy

In-situ observations on α/γ phase transformation were made to study the effects of grain boundary microstructures on the formation of a new phase and the migration of α/γ interphase boundary in an iron4. 2%Cr alloy. It was found that triple junctions with more random boundaries could be the primary nucleation sites for a new phase, while triple junctions with low angle or low ∑ coincidence boundaries did not play any role as preferential sites. The migration of α/γ interphase boundary during heating over the transformation temperature range showed the two stage behaviour characterized by a stage with a migration velocity of 0. 33-0. 75 mm/s and secondly by a stage with 3. 7-7. 6 mm/s. It was also found that abnormal grain growth and a high density of ∑3 coincidence boundaries could occur in a phase with bcc structure after cycling of α/γ phase transformation. A new mechanism of nucleation and growth of a new phase in α/γ phase transformation is proposed on the basis of roles of plane-matching interphase boundaries, as previously discussed on the origin of anisotropy of grain growth due to the migration of {110} plane-matching boundaries in Fe-3z%Si alloy. The most recent theoretical work on the distribution of plane-matching boundaries in solids with different crystal structures was found to be useful for the understanding of nucleation and growth during α/γ phase transformation.

Study on the Effect of Multi-Type Current Transformers Hybrid Operation on Differential Protection of the Bus Based on Dynamic Simulation

This paper analyzes characteristics of multi type current transformers hybrid operation for each branch of the bus and their effects on differential protection of the bus. By theoretically analyzing transmission characteristics of multi type current transformers and their influence factors, we study the dynamic model testing method of multi type current transformers for the bus, and design 3 kinds of testing schemes by making the equivalent model based on the field of P-level current transformer, TPY-level current transformer and electronic current transformer, and build the hybrid operation testing platform of multi type current transformers. Finally, we compare and analyze the transmission characteristics difference of multi type current transformers on the same branch and the characteristics difference of hybrid operation in two successive external faults, analyze the cause behind the differences, and put forward the corresponding improvement measures.

Detection of Mechanical Deformation in Old Aged Power Transformer Using Cross Correlation Co-Efficient Analysis Method

Detection of minor faults in power transformer active part is essential because minor faults may develop and lead to major faults and finally irretrievable damages occur. Sweep Frequency Response Analysis (SFRA) is an effective low-voltage, off-line diagnostic tool used for finding out any possible winding displacement or mechanical deterioration inside the Transformer, due to large electromechanical forces occurring from the fault currents or due to Transformer transportation and relocation. In this method, the frequency response of a transformer is taken both at manufacturing industry and concern site. Then both the response is compared to predict the fault taken place in active part. But in old aged transformers, the primary reference response is unavailable. So Cross Correlation Co-Efficient (CCF) measurement technique can be a vital process for fault detection in these transformers. In this paper, theoretical background of SFRA technique has been elaborated and through several case studies, the effectiveness of CCF parameter for fault detection has been represented.

Transforming growth factor-beta 1 enhances discharge activity of cortical neurons

Transforming growth factor-beta 1(TGF-β1)has been extensively studied for its pleiotropic effects on central nervous system diseases.The neuroprotective or neurotoxic effects of TGF-β1 in specific brain areas may depend on the pathological process and cell types involved.Voltage-gated sodium channels(VGSCs)are essential ion channels for the generation of action potentials in neurons,and are involved in various neuroexcitation-related diseases.However,the effects of TGF-β1 on the functional properties of VGSCs and firing properties in cortical neurons remain unclear.In this study,we investigated the effects of TGF-β1 on VGSC function and firing properties in primary cortical neurons from mice.We found that TGF-β1 increased VGSC current density in a dose-and time-dependent manner,which was attributable to the upregulation of Nav1.3 expression.Increased VGSC current density and Nav1.3 expression were significantly abolished by preincubation with inhibitors of mitogen-activated protein kinase kinase(PD98059),p38 mitogen-activated protein kinase(SB203580),and Jun NH2-terminal kinase 1/2 inhibitor(SP600125).Interestingly,TGF-β1 significantly increased the firing threshold of action potentials but did not change their firing rate in cortical neurons.These findings suggest that TGF-β1 can increase Nav1.3 expression through activation of the ERK1/2-JNK-MAPK pathway,which leads to a decrease in the firing threshold of action potentials in cortical neurons under pathological conditions.Thus,this contributes to the occurrence and progression of neuroexcitatory-related diseases of the central nervous system.

基于多尺度金字塔Transformer的人群计数方法

针对密集人群场景中背景复杂、目标尺度变化较大导致人群计数精度较低的问题,本文提出一种基于多尺度金字塔Transformer的人群计数方法(multi-scale pyramid transformer network,MSPT-Net)。在特征提取阶段设计了一种基于深度可分离自注意力的金字塔Transformer主干网络结构,该网络结构能有效捕获图像的局部和全局信息,从而有效解决人群密度图像背景复杂导致计数精度低的问题;设计了一种特征金字塔融合模块及多尺度感受野的回归头,实现了密集人群图像浅层细节特征和深层语义特征的高效融合,增强了网络对不同尺度目标的捕获能力;采用深度监督的训练方法在3个公开数据集上对提出的方法进行验证。实验结果表明,本文方法在全监督与弱监督学习策略中,与目前主流的人群计数方法相比,实现了更高精度的人群计数,克服了主流方法对背景复杂、目标尺度变化大的密集人群图像计数精度低的问题,同时本文方法保持着更小的参数量与计算量。

基于局部Transformer的泰语分词和词性标注联合模型

泰语分词和词性标注任务二者之间存在高关联性,已有研究表明将分词和词性标注任务进行联合学习可以有效提升模型性能,为此,提出了一种针对泰语拼写和构词特点的分词和词性标注联合模型。针对泰语中字符构成音节,音节组成词语的特点,采用局部Transformer网络从音节序列中学习分词特征;考虑到词根和词缀等音节与词性的关联,将用于分词的音节特征融入词语序列特征,缓解未知词的词性标注特征缺失问题。在此基础上,模型采用线性分类层预测分词标签,采用线性条件随机场建模词性序列的依赖关系。在泰语数据集LST20上的试验结果表明,模型分词F1、词性标注微平均F1和宏平均F1分别达到96.33%、97.06%和85.98%,相较基线模型分别提升了0.33%、0.44%和0.12%。

引入Transformer的道路小目标检测

针对道路场景中检测小目标时漏检率较高、检测精度低的问题,提出一种引入Transformer的道路小目标检测算法。在原YOLOv4算法基础上,对多尺度检测进行改进,把浅层特征信息充分利用起来;设计ICvT(improved convolutional vision transformer)模块捕获特征内部的相关性,获得上下文信息,提取更加全面丰富的特征;在网络特征融合部分嵌入改进后的空间金字塔池化模块,在保持较小计算量的同时增加特征图的感受野。实验结果表明,在KITTI数据集上,算法检测精度达到91.97%,与YOLOv4算法相比,mAP提高了2.53%,降低了小目标的漏检率。

一种基于Transformer的伪装目标实例分割方法

本文研究主要针对伪装复杂背景下的图像进行实例分割.由于分割对象的伪装性以及缺少大规模训练集来支撑模型训练,这对于复杂背景下的实例对象分割构成了巨大挑战.受SOTR(Segmenting Objects with Transformer)的启发下,本文设计了一种方法框架,以解决动物野生环境和其他各类复杂伪装背景下的实例分割问题.相较于原始模型在特征目标的检测表现效果不佳,以及分割掩码存在上下文信息丢失的问题,本文提出的框架主要采用以下方法:引入特定的上下文特征金字塔提取网络CEM-FPN,以解决在高分辨率图像与感受野之间的矛盾.具体而言,高分辨率图像需要更大的感受野,但大感受野会导致小目标的检测误判为背景.在多层级融合模块加入通道空间注意力模块(CS Module),以提升对不同尺度上的各个实例对象的关注度,从而提高原型掩码的质量.实验结果表明,本文方法相较于原模型在COD10K-Test数据集上提升了4.1%的精确度,并在NC4K-Test数据集上提升了4.5%的精确度.

基于Transformer-LSTM的闽南语唇语识别

针对端到端句子级闽南语唇语识别的问题,提出一种基于Transformer和长短时记忆网络(LSTM)的编解码模型.编码器采用时空卷积神经网络及Transformer编码器用于提取唇读序列时空特征,解码器采用长短时记忆网络并结合交叉注意力机制用于文本序列预测.最后,在自建闽南语唇语数据集上进行实验.实验结果表明:模型能有效地提高唇语识别的准确率.

Optimal Economic Life Interval Evaluation Method of Transformers

This paper proposed a method of optimal economic life interval evaluation of transformer, thus providing scientific support for power system planning and reconstruction. First, obtaining the transformer failure rate by using Levenberg-Marquardt based least squares algorithm on the consumption that transformer failure mode meets three-parameters Weibull distribution. Second, calculating operation cost and failure cost of transformer during its life cycle on the basis of transformer failure rate. Then, acquiring the optimal economic life interval of transformer through life cycle minimal average annual cost based on the theory of interval analysis. Finally, the proposed method is proved to be effective by applying to some real transformers.

基于Transformer的多编码器端到端语音识别

当前广泛使用的Transformer模型具有良好的全局依赖关系捕捉能力,但其在浅层时容易忽略局部特征信息。针对该问题,文中提出了一种使用多个编码器来改善语音特征信息提取能力的方法。通过附加一个额外的卷积编码器分支来强化对局部特征信息的捕捉,弥补浅层Transformer对局部特征信息的忽视,有效实现音频特征序列全局和局部依赖关系的融合,即提出了基于Transformer的多编码器模型。在开源中文普通话数据集Aishell-1上的实验表明,在没有外部语言模型的情况下,相比于Transformer模型,基于Transformer的多编码器模型的字符错误率降低了4.00%。在内部非公开的上海话方言数据集上,文中所提模型的性能提升更加明显,其字符错误率从19.92%降低至10.31%,降低了48.24%。