基于Group-Res2Block的智能合成语音说话人确认方法

针对现有说话人确认任务基于自然语音条件下并不适用于智能合成语音的问题,提出一种基于Group-Res2Block的智能合成语音说话人确认方法。首先,设计了Group-Res2Block结构,在Res2Block的基础上将当前分组与相邻前后分组进行合并形成新的分组,以增强说话人局部特征的上下文联系;其次,设计了并行结构的多尺度通道注意力特征融合机制,利用不同大小卷积核实现同一层级的特征在通道维度的特征选择,以获取更具表现力的说话人特征,避免信息冗余;最后,设计了串行结构的多尺度层注意力特征融合机制,构建层结构,将深浅层特征整体进行融合并赋予不同权重,以获取最优的特征表达。为验证所提出特征提取网络的有效性,构建了中英文两种智能合成语音数据集进行消融实验和对比实验。结果表明本文方法在该任务的评价指标精确度(accuracy,ACC)、等错误率(equal error rate,EER)和最小检测代价函数(minimum detection cost function,minDCF)上是最优的。此外,通过对模型泛化性能进行测试,验证了本文方法对未知智能语音算法的适用性。

Bone marrow-derived mesenchymal stem cell-derived exosomeloaded miR-129-5p targets high-mobility group box 1 attenuates neurological-impairment after diabetic cerebral hemorrhage

BACKGROUND Diabetic intracerebral hemorrhage(ICH)is a serious complication of diabetes.The role and mechanism of bone marrow mesenchymal stem cell(BMSC)-derived exosomes(BMSC-exo)in neuroinflammation post-ICH in patients with diabetes are unknown.In this study,we investigated the regulation of BMSC-exo on hyperglycemia-induced neuroinflammation.AIM To study the mechanism of BMSC-exo on nerve function damage after diabetes complicated with cerebral hemorrhage.METHODS BMSC-exo were isolated from mouse BMSC media.This was followed by transfection with microRNA-129-5p(miR-129-5p).BMSC-exo or miR-129-5poverexpressing BMSC-exo were intravitreally injected into a diabetes mouse model with ICH for in vivo analyses and were cocultured with high glucoseaffected BV2 cells for in vitro analyses.The dual luciferase test and RNA immunoprecipitation test verified the targeted binding relationship between miR-129-5p and high-mobility group box 1(HMGB1).Quantitative polymerase chain reaction,western blotting,and enzyme-linked immunosorbent assay were conducted to assess the levels of some inflammation factors,such as HMGB1,interleukin 6,interleukin 1β,toll-like receptor 4,and tumor necrosis factorα.Brain water content,neural function deficit score,and Evans blue were used to measure the neural function of mice.RESULTS Our findings indicated that BMSC-exo can promote neuroinflammation and functional recovery.MicroRNA chip analysis of BMSC-exo identified miR-129-5p as the specific microRNA with a protective role in neuroinflammation.Overexpression of miR-129-5p in BMSC-exo reduced the inflammatory response and neurological impairment in comorbid diabetes and ICH cases.Furthermore,we found that miR-129-5p had a targeted binding relationship with HMGB1 mRNA.CONCLUSION We demonstrated that BMSC-exo can reduce the inflammatory response after ICH with diabetes,thereby improving the neurological function of the brain.

Grid Side Distributed Energy Storage Cloud Group End Region Hierarchical Time-Sharing Configuration Algorithm Based onMulti-Scale and Multi Feature Convolution Neural Network

There is instability in the distributed energy storage cloud group end region on the power grid side.In order to avoid large-scale fluctuating charging and discharging in the power grid environment and make the capacitor components showa continuous and stable charging and discharging state,a hierarchical time-sharing configuration algorithm of distributed energy storage cloud group end region on the power grid side based on multi-scale and multi feature convolution neural network is proposed.Firstly,a voltage stability analysis model based onmulti-scale and multi feature convolution neural network is constructed,and the multi-scale and multi feature convolution neural network is optimized based on Self-OrganizingMaps(SOM)algorithm to analyze the voltage stability of the cloud group end region of distributed energy storage on the grid side under the framework of credibility.According to the optimal scheduling objectives and network size,the distributed robust optimal configuration control model is solved under the framework of coordinated optimal scheduling at multiple time scales;Finally,the time series characteristics of regional power grid load and distributed generation are analyzed.According to the regional hierarchical time-sharing configuration model of“cloud”,“group”and“end”layer,the grid side distributed energy storage cloud group end regional hierarchical time-sharing configuration algorithm is realized.The experimental results show that after applying this algorithm,the best grid side distributed energy storage configuration scheme can be determined,and the stability of grid side distributed energy storage cloud group end region layered timesharing configuration can be improved.

Multi-Criteria Group Decision-Making Method Based on an Improved Single-Valued Neutrosophic Hamacher Weighted Average Operator and Grey Relational Analysis

This paper proposes a multi-criteria decision-making (MCGDM) method based on the improved single-valued neutrosophic Hamacher weighted averaging (ISNHWA) operator and grey relational analysis (GRA) to overcome the limitations of present methods based on aggregation operators. First, the limitations of several existing single-valued neutrosophic weighted averaging aggregation operators (i.e. , the single-valued neutrosophic weighted averaging, single-valued neutrosophic weighted algebraic averaging, single-valued neutrosophic weighted Einstein averaging, single-valued neutrosophic Frank weighted averaging, and single-valued neutrosophic Hamacher weighted averaging operators), which can produce some indeterminate terms in the aggregation process, are discussed. Second, an ISNHWA operator was developed to overcome the limitations of existing operators. Third, the properties of the proposed operator, including idempotency, boundedness, monotonicity, and commutativity, were analyzed. Application examples confirmed that the ISNHWA operator and the proposed MCGDM method are rational and effective. The proposed improved ISNHWA operator and MCGDM method can overcome the indeterminate results in some special cases in existing single-valued neutrosophic weighted averaging aggregation operators and MCGDM methods.

基于三重博弈的多微网-产消者群自适应鲁棒分布式协同优化

在多主体能源共享背景下,提出一种基于三重博弈的多微网-产消者群自适应鲁棒分布式协同优化。首先,构建三重博弈框架模拟多微网-产消者群的能源交易模型,即上层多微网系统合作博弈,各微网与其下层产消者群主从博弈,下层产消者群联盟合作博弈。其次,考虑到多微网系统内部受可再生能源出力、负荷功率随机性,外部受日前电价波动性等多重不确定性影响较大,各微网建立两阶段自适应鲁棒优化(adaptive robust optimization,ARO)模型,在第一阶段优化微网与外部主体的交易策略,包括与其他微网的合作博弈、与产消者群的主从博弈以及与配网日前电价的不确定性;在第二阶段基于多面体不确定集来决策微网内部源荷的不确定性,优化本地机组的调度策略。然后,对于各层合作博弈收益,采用改进纳什谈判模型对其进行分配。最后,采用一种可变惩罚参数交替方向乘子法-列与约束生成算法(variable penalty parameters alternating direction multiplier method-column and constraint generation,VPPADMM-C&CG)分布式求解三重博弈ARO模型,以保护各主体的隐私。通过不同博弈场景的算例结果对所提模型能够实现多微网-产消者群社会总成本最小化及效益最大化进行了验证,可为不确定环境下多微网-产消者群协同优化提供参考。

基于GRA-GCRITIC和改进加权双向投影的区间Fermatean模糊多属性群决策方法

针对司法案件决策环境的复杂性及评价信息的模糊性,提出一种基于GRA-GCRITIC(grey relational analysis-group criteria importance through intercriteria correlation)和改进加权双向投影的区间Fermatean模糊Hamacher-TODIM多属性群决策方法。首先考虑到各位专家对各评价方案下各指标评价信息的差异性,提出一种GRA-GCRITIC方法,该方法将灰色关联融入到CRITIC中,以确定指标综合权重比单一获取的指标权重更加客观可靠;其次,结合信任关系对加权双向投影法进行改进,兼顾主客观关系,利用专家个体与群体评价信息的隶属度及非隶属度矩阵间的相似度得到专家权重;最后,考虑到决策者的损失规避心理,将融合了Hamacher算子的TODIM方法拓展至区间Fermatean模糊环境中,通过具体算例可得到其综合优势度及排序,验证了所提方法的可行性及灵活性。

基于CMCP-LMCL的多分类深度神经网络及其应用

多分类问题涉及信用风险管理、股票走势预测等多个领域。深度神经网络(DNN)是常用于多分类预测的机器学习模型,然而输入特征维度较高且存在冗余信号时,将加重其可解释性不强和结构冗余等缺陷;同时,常用的Softmax损失也可能面临分类边界模糊导致预测效果不佳等问题。为此,本文针对多分类问题,提出一个新的深度神经网络CMCP-LMCL,利用CMCP变量选择方法压缩输入特征到第1隐藏层的权重。该方法融合权重的组结构,能够剔除无关特征以及不重要的连接;同时,对特征层之外的权重施加权重衰减L;2;惩罚,有利于改进过拟合问题。新方法的增强边缘余弦损失(LMCL)在Softmax基础上引入扩大参数和距离参数,增大分类决策边界的间隔以期提高分类预测性能。模拟分析表明,对比已有DNN和传统分类方法,无论特征以简单线性形式还是复杂非线性形式映射到因变量,本文所提出的方法均具有良好的特征选择性能和预测表现。基于信用贷款数据的实证分析表明,该方法能够有效选择风险指标并进行违约风险预警。

基于决策者偏好构造的概率语言型Z-number多属性决策方法

由于决策环境的日益复杂,专家评价信息往往会包含专家偏好,因此有必要对专家偏好进行识别与处理。针对这一情况,提出了基于决策者偏好构造的概率语言型Z-number并应用于多属性决策问题中。首先,针对专家偏好未知的情况,提出了一种基于修正的平均得分距离,由专家评价信息确定专家偏好。然后,依据得到的专家偏好,利用GOWPA算子与最大熵生成概率语言型Z-number。同时,获得专家评价信息的可信度,由专家可信度聚合专家信息并采用TODIM方法对方案进行排序。最后,在国家大力提倡节能减排的背景下,将所提方法应用于国家绿色矿山试点企业的选取上,验证了所提方法的有效性和可行性。

计及群体压力和群体极化的PDHFSs-TODIM多属性群体决策算法

在概率对偶犹豫模糊环境下的多属性决策中,针对决策过程一般未考虑决策专家在讨论中所受的群体压力和群体极化的问题,提出了一种计及群体压力和群体极化的概率对偶犹豫模糊集-拓扑判别目标间方法(probabilistic discriminative heterogeneous fuzzy sets-topological discriminative inter-objective method,PDHFSsTODIM)的多属性群体决策算法。首先,定义了考虑群体压力的决策信息修正公式,有效削弱群体压力对专家决策信息的影响;然后,搭建了一种考虑群体极化现象的决策模型,该模型充分考虑了专家的主观性,为专家讨论和修正决策提供了机会,通过分配动态决策权重提高决策的置信度,并通过群体极化阈值约束群体的极化效应;进一步,将计算得到的决策数据输入改进的TODIM算法,得到最终决策结果。将算法运用于储能电池综合性能评估分析,通过与其他先进决策算法做对比,验证了所提算法的可靠性与准确性。

GLOBAL STABILITY OF EXTENDED MULTI-GROUP SIR EPIDEMIC MODELS WITH PATCHES THROUGH MIGRATION AND CROSS PATCH INFECTION

In this article, we establish the global stability of an endemic equilibrium of multi-group SIR epidemic models, which have not only an exchange of individuals between patches through migration but also cross patch infection between different groups. As a result, we partially generalize the recent result in the article [16].

A modified multi-group model of angular and momentum distribution of cosmic ray muons for thickness measurement and material discrimination of slabs

Muon tomography is a capable imaging technique to measure the geometry of high-Z objects. However,most existed algorithms used in muon tomography have obscured the effects of angular distribution and momentum spectra of cosmic ray muons and reduced the spatial resolution. We present a modified multi-group model that takes into account these effects and calibrates the model by the material of lead. Performance tests establish that the model is capable of measuring the thickness of a Pb slab and identifying the material of an unknown slab on a reasonable exposure timescale, in both cases of complete and incomplete angular data. Results show that the modified multi-group model is helpful for improvements in image resolution in real applications.

Novel combinatorial algorithm for the problems of fuzzy grey multi-attribute group decision making

To study the fuzzy and grey information in the problems of multi-attribute group decision making, the basic concepts of both fuzzy grey numbers and grey interval numbers are given firstly, then a new model of fuzzy grey multi-attribute group decision making based on the theories of fuzzy mathematics and grey system is presented. Furthermore, the grey interval relative degree and deviation degree is defined, and both the optimistic algorithm of the grey interval relational degree and the algorithm of deviation degree minimization for solving this new model are also given. Finally, a decision making example to demonstrate the feasibility and rationality of this new method is given, and the results by using these two algorithms are uniform.

Global dynamics of a novel multi-group model for computer worms

In this paper, we study worm dynamics in computer networks composed of many autonomous systems. A novel multigroup SIQR （susceptible-infected-quarantined-removed） model is proposed for computer worms by explicitly considering anti-virus measures and the network infrastructure. Then, the basic reproduction number of worm R0 is derived and the global dynamics of the model are established. It is shown that if R0 is less than or equal to 1, the disease-free equilibrium is globally asymptotically stable and the worm dies out eventually, whereas, if R0 is greater than 1, one unique endemic equilibrium exists and it is globally asymptotically stable, thus the worm persists in the network. Finally, numerical simulations are given to illustrate the theoretical results.

Multi-group ant colony algorithm based on simulated annealing method

To overcome the default of single search tendency, the ants in the colony are divided into several sub-groups. The ants in different subgroups have different trail information and expectation coefficients. The simulated annealing method is introduced to the algorithm. Through setting the temperature changing with the iterations, after each turn of tours, the solution set obtained by the ants is taken as the candidate set. The update set is obtained by adding the solutions in the candidate set to the previous update set with the probability determined by the temperature. The solutions in the candidate set are used to update the trail information. In each turn of updating, the current best solution is also used to enhance the trail information on the current best route. The trail information is reset when the algorithm is in stagnation state. The computer experiments demonstrate that the proposed algorithm has higher stability and convergence speed.

Multi-group pin power reconstruction method based on colorset form functions

A multi-group pin power reconstruction method that fully exploits nodal information obtained from global coarse mesh solution has been developed.It expands the intra-nodal flux distributions into nonseparable semi-analytic basis functions,and a colorset based form function generating method is proposed,which can accurately model the spectral interaction occurring at assembly interface.To demonstrate its accuracy and applicability to realistic problems, the new method is tested against two benchmark problems,including a mixed-oxide fuel problem.The results show that the new method is comparable in accuracy to fine-mesh methods.

Multi-proxy quantum group signature scheme with threshold shared verification

A multi-proxy quantum group signature scheme with threshold shared verification is proposed. An original signer may authorize a proxy group as his proxy agent. Then only the cooperation of all the signers in the proxy group can generate the proxy signature on behalf of the original signer. In the scheme, any t or more of n receivers can verify the message and any t - 1 or fewer receivers cannot verify the validity of the proxy signature.

Multi-criteria group decision making with fuzzy data:an extension of the VIKOR method

The VIKOR method is a multi-criteria decision making aid, which employs linear normalization to offer compromise solu- tions and has been successfully applied to various group decision making problems. However, the conventional VIKOR techniques used to integrate group judgments and the information loss arising from defuzzification are problematic and distort final outcomes. An improved integration method, which is optimization-based, is proposed. And it can handle fuzzy criteria values and weights. The precondition for accurately defuzzifying triangular fuzzy num- bers is identified. Several effective defuzzification procedures are proposed to improve the extant VIKOR, and a comprehensive evaluation framework is offered to aid multi-criteria group decision making. Finally, a numerical example is provided to illustrate the practicability of the proposed method.

Power Aggregation Operators of Simplified Neutrosophic Sets and Their Use in Multi-attribute Group Decision Making

The simplified neutrosophic set(SNS) is a useful generalization of the fuzzy set that is designed for some practical situations in which each element has different truth membership function, indeterminacy membership function and falsity membership function. In this paper, we develop a series of power aggregation operators called simplified neutrosophic number power weighted averaging(SNNPWA) operator, simplified neutrosophic number power weighted geometric(SNNPWG) operator, simplified neutrosophic number power ordered weighted averaging(SNNPOWA) operator and simplified neutrosophic number power ordered weighted geometric(SNNPOWG) operator. We present some useful properties of the operators and discuss the relationships among them. Moreover, an approach to multiattribute group decision making(MAGDM) within the framework of SNSs is developed by the above aggregation operators.Finally, a practical application of the developed approach to deal with the problem of investment is given, and the result shows that our approach is reasonable and effective in dealing with uncertain decision making problems.

Dynamic programming methodology for multi-criteria group decision-making under ordinal preferences

A method of minimizing rankings inconsistency is proposed for a decision-making problem with rankings of alternatives given by multiple decision makers according to multiple criteria. For each criteria, at first, the total inconsistency between the rankings of all alternatives for the group and the ones for every decision maker is defined after the decision maker weights in respect to the criteria are considered. Similarly, the total inconsistency between their final rankings for the group and the ones under every criteria is determined after the criteria weights are taken into account. Then two nonlinear integer programming models minimizing respectively the two total inconsistencies above are developed and then transformed to two dynamic programming models to obtain separately the rankings of all alternatives for the group with respect to each criteria and their final rankings. A supplier selection case illustrated the proposed method, and some discussions on the results verified its effectiveness. This work develops a new measurement of ordinal preferences’ inconsistency in multi-criteria group decision-making （MCGDM） and extends the cook-seiford social selection function to MCGDM considering weights of criteria and decision makers and can obtain unique ranking result.

Pythagorean Uncertain Linguistic Variable Hamy Mean Operator and Its Application to Multi-attribute Group Decision Making

Pythagorean fuzzy set(PFS) can provide more flexibility than intuitionistic fuzzy set(IFS) for handling uncertain information, and PFS has been increasingly used in multi-attribute decision making problems. This paper proposes a new multiattribute group decision making method based on Pythagorean uncertain linguistic variable Hamy mean(PULVHM) operator and VIKOR method. Firstly, we define operation rules and a new aggregation operator of Pythagorean uncertain linguistic variable(PULV) and explore some properties of the operator.Secondly, taking the decision makers' hesitation degree into account, a new score function is defined, and we further develop a new group decision making approach integrated with VIKOR method. Finally, an investment example is demonstrated to elaborate the validity of the proposed method. Sensibility analysis and comprehensive comparisons with another two methods are performed to show the stability and advantage of our method.