A Two-Layer Encoding Learning Swarm Optimizer Based on Frequent Itemsets for Sparse Large-Scale Multi-Objective Optimization

Traditional large-scale multi-objective optimization algorithms(LSMOEAs)encounter difficulties when dealing with sparse large-scale multi-objective optimization problems(SLM-OPs)where most decision variables are zero.As a result,many algorithms use a two-layer encoding approach to optimize binary variable Mask and real variable Dec separately.Nevertheless,existing optimizers often focus on locating non-zero variable posi-tions to optimize the binary variables Mask.However,approxi-mating the sparse distribution of real Pareto optimal solutions does not necessarily mean that the objective function is optimized.In data mining,it is common to mine frequent itemsets appear-ing together in a dataset to reveal the correlation between data.Inspired by this,we propose a novel two-layer encoding learning swarm optimizer based on frequent itemsets(TELSO)to address these SLMOPs.TELSO mined the frequent terms of multiple particles with better target values to find mask combinations that can obtain better objective values for fast convergence.Experi-mental results on five real-world problems and eight benchmark sets demonstrate that TELSO outperforms existing state-of-the-art sparse large-scale multi-objective evolutionary algorithms(SLMOEAs)in terms of performance and convergence speed.

Novel joint encoding/decoding algorithms of fountain codes for underwater acoustic communication

Fountain codes are considered to be a promising coding technique in underwater acoustic communication（UAC） which is challenged with the unique propagation features of the underwater acoustic channel and the harsh marine environment. And Luby transform（LT） codes are the first codes fully realizing the digital fountain concept. However, in conventional LT encoding/decoding algorithms, due to the imperfect coverage（IC） of input symbols and short cycles in the generator matrix, stopping sets would occur and terminate the decoding. Thus, the recovery probability is reduced,high coding overhead is required and decoding delay is increased.These issues would be disadvantages while applying LT codes in underwater acoustic communication. Aimed at solving those issues, novel encoding/decoding algorithms are proposed. First,a doping and non-uniform selecting（DNS） encoding algorithm is proposed to solve the IC and the generation of short cycles problems. And this can reduce the probability of stopping sets occur during decoding. Second, a hybrid on the fly Gaussian elimination and belief propagation（OFG-BP） decoding algorithm is designed to reduce the decoding delay and efficiently utilize the information of stopping sets. Comparisons via Monte Carlo simulation confirm that the proposed schemes could achieve better overall decoding performances in comparison with conventional schemes.

Efficient Lubytransform encodingalgorithm based on short cycle elimination

An effective Luby transform （LT） encoding algorithm based on short cycle elimination is proposed to improve decoding probabilities of short length LT codes. By searching the generator ma- trix, some special encoded symbols are generated by the encoder to effectively break the short cycles that have negative effect on the performance of LT codes. Analysis and numerical results show that by employing the proposed algorithm, the encoding complexity decreases and the decoding probabili- ties improve both in binary erasure channels （BECs） and additive white gauss noise （AWGN） chan- nels.

Permutation Encoding for Pilot Coordination in Multi-user Massive MIMO

Pilot plays an essential role in a duplex communication system.Several methods have been proposed for pilot assignment over specific scenarios.With the help of permutation encoding,we implemented a genetic algorithm for optimizing pilot assignment in a multi-user massive multiple input multiple output(MIMO)system.Results show improvement on existing results especially in the case of strong user estimation rates.

Pseudo-Random Coding Side-Lobe Suppression Method Based on CLEAN Algorithm

A pseudo-random coding side-lobe suppression method based on CLEAN algorithm is introduced.The CLEAN algorithm mainly processes pulse compression results of a pseudo-random coding,and estimates a target＇s distance by a method named interpolation method,so that we can get an ideal pulse compression result of the target,and then use the adjusted ideal pulse compression side-lobe to cut the actual pulse compression result,so as to achieve the remarkable performance of side-lobe suppression for large targets,and let the adjacent small targets appear.The computer simulations by MATLAB with this method analyze the effect of side-lobe suppression in an ideal or noisy environment.It is proved that this method can effectively solve the problem due to the side-lobe of pseudo-random coding being too high,and can enhance the radar＇s multi-target detection ability.

Quantum-Inspired Particle Swarm Optimization Algorithm Encoded by Probability Amplitudes of Multi-Qubits

To enhance the optimization ability of particle swarm algorithm, a novel quantum-inspired particle swarm optimization algorithm is proposed. In this method, the particles are encoded by the probability amplitudes of the basic states of the multi-qubits system. The rotation angles of multi-qubits are determined based on the local optimum particle and the global optimal particle, and the multi-qubits rotation gates are employed to update the particles. At each of iteration, updating any qubit can lead to updating all probability amplitudes of the corresponding particle. The experimental results of some benchmark functions optimization show that, although its single step iteration consumes long time, the optimization ability of the proposed method is significantly higher than other similar algorithms.

ASP源代码加密程序Script Encoder算法研究

微软为了保护脚本代码的安全性 ,以 COM组件的形式提供了一种对脚本代码进行编码加密的技术 .但其安全性到底如何呢 ?本文将通过“反编译”的方式对其加密。

面向不间断供电(UPS)系统的电能质量分析技术

针对现行UPS电能质量检测过程存在的准确率低、实时性差且成本高的缺点,文中基于VMD-SAE-1DCNN模型提出了一种UPS电能质量检测与识别算法。对于电能信号非线性与非平稳的特点,使用变分模态算法对原信号进行分解,从而得到本征模态信号。同时,采用稀疏自编码器对本征模态信号进行特征提取,通过建立多层一维卷积神经网络模型对特征进行训练,提升了运算效率。实验测试结果表明,所提算法的迭代次数与运行时间在对比算法中均为最优,分类准确率可达97%以上,充分证明了改进算法的有效性。

GLOBAL OPTIMIZATION OF PUMP CONFIGURATION PROBLEM USING EXTENDED CROWDING GENETIC ALGORITHM

An extended crowding genetic algorithm (ECGA) is introduced for solvingoptimal pump configuration problem, which was presented by T. Westerlund in 1994. This problem hasbeen found to be non-convex, and the objective function contained several local optima and globaloptimality could not be ensured by all the traditional MINLP optimization method. The concepts ofspecies conserving and composite encoding are introduced to crowding genetic algorithm (CGA) formaintain the diversity of population more effectively and coping with the continuous and/or discretevariables in MINLP problem. The solution of three-levels pump configuration got from DICOPT++software (OA algorithm) is also given. By comparing with the solutions obtained from DICOPT++, ECPmethod, and MIN-MIN method, the ECGA algorithm proved to be very effective in finding the globaloptimal solution of multi-levels pump configuration via using the problem-specific information.

A genetic algorithm for community detection in complex networks

A new genetic algorithm for community detection in complex networks was proposed. It adopts matrix encoding that enables traditional crossover between individuals. Initial populations are generated using nodes similarity, which enhances the diversity of initial individuals while retaining an acceptable level of accuracy, and improves the efficiency of optimal solution search. Individual crossover is based on the quality of individuals' genes; all nodes unassigned to any community are grouped into a new community, while ambiguously placed nodes are assigned to the community to which most of their neighbors belong. Individual mutation, which splits a gene into two new genes or randomly fuses it into other genes, is non-uniform. The simplicity and effectiveness of the algorithm are revealed in experimental tests using artificial random networks and real networks. The accuracy of the algorithm is superior to that of some classic algorithms, and is comparable to that of some recent high-precision algorithms.

Application of a Genetic Algorithm Based on the Immunity for Flow Shop under Uncertainty

The uncertain duration of each job in each machine in flow shop problem was regarded as an independent random variable and was described by mathematical expectation.And then,an immune based partheno-genetic algorithm was proposed by making use of concepts and principles introduced from immune system and genetic system in nature.In this method,processing se- quence of products could be expressed by the character encoding and each antibody represents a feasible schedule.Affinity was used to measure the matching degree between antibody and antigen.Then several antibodies producing operators,such as swopping,mov- ing,inverting,etc,were worked out.This algorithm was combined with evolution function of the genetic algorithm and density mechanism in organisms immune system.Promotion and inhibition of antibodies were realized by expected propagation ratio of an- tibodies,and in this way,premature convergence was improved.The simulation proved that this algorithm is effective.

An Improved Genetic Algorithm for Solving the Mixed⁃Flow Job⁃Shop Scheduling Problem with Combined Processing Constraints

The flexible job-shop scheduling problem(FJSP)with combined processing constraints is a common scheduling problem in mixed-flow production lines.However,traditional methods for classic FJSP cannot be directly applied.Targeting this problem,the process state model of a mixed-flow production line is analyzed.On this basis,a mathematical model of a mixed-flow job-shop scheduling problem with combined processing constraints is established based on the traditional FJSP.Then,an improved genetic algorithm with multi-segment encoding,crossover,and mutation is proposed for the mixed-flow production line problem.Finally,the proposed algorithm is applied to the production workshop of missile structural components at an aerospace institute to verify its feasibility and effectiveness.

Neural network fault diagnosis method optimization with rough set and genetic algorithms

Aiming at the disadvantages of BP model in artificial neural networks applied to intelligent fault diagnosis, neural network fault diagnosis optimization method with rough sets and genetic algorithms are presented. The neural network nodes of the input layer can be calculated and simplified through rough sets theory; The neural network nodes of the middle layer are designed through genetic algorithms training; the neural network bottom-up weights and bias are obtained finally through the combination of genetic algorithms and BP algorithms. The analysis in this paper illustrates that the optimization method can improve the performance of the neural network fault diagnosis method greatly.

Application of Improved Deep Auto-Encoder Network in Rolling Bearing Fault Diagnosis

Since the effectiveness of extracting fault features is not high under traditional bearing fault diagnosis method, a bearing fault diagnosis method based on Deep Auto-encoder Network (DAEN) optimized by Cloud Adaptive Particle Swarm Optimization (CAPSO) was proposed. On the basis of analyzing CAPSO and DAEN, the CAPSO-DAEN fault diagnosis model is built. The model uses the randomness and stability of CAPSO algorithm to optimize the connection weight of DAEN, to reduce the constraints on the weights and extract fault features adaptively. Finally, efficient and accurate fault diagnosis can be implemented with the Softmax classifier. The results of test show that the proposed method has higher diagnostic accuracy and more stable diagnosis results than those based on the DAEN, Support Vector Machine (SVM) and the Back Propagation algorithm (BP) under appropriate parameters.

Two-stage evolutionary algorithm for dynamic multicast routing in mesh network

In order to share multimedia transmissions in mesh networks and optimize the utilization of network resources, this paper presents a Two-stage Evolutionary Algorithm (TEA), i.e., unicast routing evolution and multicast path composition, for dynamic multicast routing. The TEA uses a novel link-duplicate-degree encoding, which can encode a multicast path in the link-duplicate-degree and decode the path as a link vector easily. A dynamic algorithm for adding nodes to or removing nodes from a multicast group and a repairing algorithm are also covered in this paper. As the TEA is based on global evaluation, the quality of the multicast path remains stabilized without degradation when multicast members change over time. Therefore, it is not necessary to rearrange the multicast path during the life cycle of the multicast sessions. Simulation results show that the TEA is efficient and convergent.

联合训练下融合编解码器的序列推荐算法

现有基于Transformer的推荐算法通常仅考虑使用编码器来进行推荐预测,缺乏利用解码器去“解码”用户行为序列的能力,不能较为准确预测用户下一次的交互行为。为解决此问题,基于阿里巴巴电子商务推荐的行为序列模型(BST)提出联合训练下融合编解码器的序列推荐算法模型BSTEAD。通过采用联合训练机制,设置Transformer预测任务和BST预测任务。将两条预测任务的损失进行加权求和,得到最终的损失函数。在MovieLens和Goodbooks两个公共数据集上的实验结果表明,BSTEAD推荐算法与5个对比模型相比性能具有显著提升,验证了联合训练机制下解码器对推荐任务的有效性。

高效前缀约简的三维Hilbert空间填充曲线编解码算法

3维Hilbert空间填充曲线(3D HSFC)的编码和解码效率对空间查询处理、图像处理等领域的应用举足轻重。现有的3维编解码算法独立编解码每一个点,忽略了Hilbert曲线的局部保持特性。为了提高编解码效率,该文设计了高效的3D状态视图,并提出一种新的前缀约简的3D HSFC编码算法(PR-3HE)和前缀约简3D HSFC解码算法(PR-3HD),这两个算法通过公共前缀的定义和识别、公共前缀约简及多种优化技术来最小化需要编码的阶数,从而提高3D HSFC的编解码效率。理论上证明:当编码或解码一个k阶的窗体(窗体内总共含有2k×2k×2k个点)时,PR-3HE平均每个点的编码阶数不超过2,PR-3HD平均解码阶数不超过8/7。相对于传统的基于迭代的方法,编解码时间复杂度从O(k)降低到了O(1)。实验结果表明,该文算法在模拟数据集和真实数据集上的表现显著优于现有算法。

基于稀疏自编码的多维数据去重聚类算法分析

随着科技信息的不断发展,数据量与数据类型与日俱增,针对数据集维度高、重复数据多导致有效信息提取复杂的问题,提出基于改进稀疏自编码器的多维数据聚类算法。算法分为数据处理与聚类分析两大部分,数据处理时首先利用S-SAE中逐层贪婪的原理将高维数据集降维至每组6维的数据集;接着采用映射值匹配机制对降维后的数据集进行重复数据清洗处理,被清洗的值用0替代;然后将处理好的数据投入到K-Means++聚类算法中进行聚类分析;最终构建出TS-SAE-K-Means++多维数据聚类模型,并通过最优化分析得出其最优化参数设置情况。通过对不同基线组合算法的仿真对比分析表明,TS-SAE-K-Means++在聚类轮廓系数S与模型特征值F1评价体系中均优于其它算法组合。这表明提出的算法在解决高维数据内有效信息提取的问题上具有一定的优越性。

一种POA-VMD和自编码器结合的风电机组轴承劣化指标构建及故障诊断方法

针对目前轴承性能劣化指标的构建及故障诊断高度依赖专家经验,限制条件繁多,实际应用情景单一的问题,提出一种鹈鹕优化算法(POA)优化的变分模态分解(VMD)和自编码器结合的风机轴承劣化指标构建及故障诊断方法。首先利用POA-VMD算法将轴承振动信号采用自适应方法分解为K个固有模态分量(IMF),并针对上述分量分别构建K个自编码器;然后以正常状态振动信号的分解结果为训练样本完成自编码器的训练,并以训练完成后模型的输出结果为基础构建轴承劣化指标,借助劣化指标监测轴承早期微弱故障;最后对故障时刻振动信号的IMF分量重构结果进行包络谱分析,确定故障的类型。经实验验证:该方法不仅可以清晰地展现轴承的劣化过程,对早期微弱故障敏感性高,而且在故障发生后可以准确诊断出故障类型。