上下文分解机的自适应更新策略

分解机模型已经被成功应用于上下文推荐系统。在分解机模型的学习算法中,交替最小二乘法是一种固定其他参数只求单一参数最优值的学习算法,其参数数目影响计算复杂度。然而当特征数目很大时,参数数目随着特征数目急剧增加,导致计算复杂度很高;即使有些参数已经达到了最优值,每次迭代仍更新所有的参数。因此,主要改进了交替最小二乘法的参数更新策略,为参数引入自适应误差指标,通过权重和参数绝对误差共同决定该参数更新与否,使得每次迭代时重点更新最近两次迭代取值变化较大的参数。这种仅更新自适应误差大于阈值的参数的策略不但减少了需要更新的参数数目,进而加快了算法收敛的速度和缩短了运行时间,而且参数权重由误差决定,又修正了误差。在Yahoo和Movielens数据集上的实验结果证明:改进的参数更新策略运行效率有明显提高。     

An efficient stabilized fast Newton adaptive filtering algorithm for stereophonic acoustic echo cancellation SAEC

This paper addresses the field of stereophonic acoustic echo cancellation (SAEC) by adaptive filtering algorithms. Recently, we have proposed a new version of the fast Newton transversal FNTF algorithm for SAEC applications. In this paper, we propose an efficient modification of this algorithm for the same applications. This new algorithm uses a new proposed and simplified numerical stabilization technique and takes into account the cross-correlation between the inputs of the channels. The basic idea is to introduce a small nonlinearity into each channel that has the effect of reducing the inter-channel coherence while not being noticeable for speech due to self masking. The complexity of the proposed algorithm does not alter the complexity of the original version and is kept less than half the complexity of the fastest two-channel FTF filter version. Simulation results and comparisons with the extended two-channel normalized least mean square NLMS and FTF algorithms are presented.     

基于逐次超松弛技术的Double Speedy Q-Learning算法

Q-Learning是目前一种主流的强化学习算法,但其在随机环境中收敛速度不佳,之前的研究针对Speedy Q-Learning存在的过估计问题进行改进,提出了Double Speedy Q-Learning算法.但Double Speedy Q-Learning算法并未考虑随机环境中存在的自循环结构,即代理执行动作时...     

一种新的分割式比例仿射投影算法

针对自适应算法收敛速度和计算复杂度之间的矛盾．提出一种基于集员滤波的分割式比例仿射投影算法（SM-SPAPA）。该算法中只有当参数估计误差大于给定的误差门限时滤波器系数才进行迭代更新,从而能有效地减少滤波器系数的迭代次数。仿真结果表明,由于每次迭代将对误差性能贡献最大的输入信号筛选出来作为输入,从而能加快收敛速度,同时还能够减少算法的运算量。     

混合粒子群和差分进化的定位算法

针对由测量误差造成的无线传感器网络定位精度不高的问题,提出一种混合粒子群和差分进化的节点定位算法（HPSO-DE）。首先,对粒子群算法的惯性权重进行自适应更新,使得每个个体随着迭代次数的增加而增大,进而提高其全局探索能力,然后改进差分进化算法的变异策略,从而提高该算法的局部寻优能力,之后将个体先经过改进的粒子群算法优化,低于平均适应度值的个体继续通过改进的差分进化算法优化,从而得到HPSO-DE算法。HPSO-DE算法继承了二者的优点,提高了该算法的最优解精度和收敛速度。最后在无线传感器网络节点定位模型中应用HPSO-DE算法,仿真结果表明,所提HPSO-DE算法在测距误差为30%时,定位误差比PSO和DFOA分别少2.1m和1.1m,具有更高的定位精度和更强的抗误差性能。     

A low-complexity hybrid algorithm based on particle swarm and ant colony optimization for large-MIMO detection

With rapid increase in demand for higher data rates, multiple-input multiple-output (MIMO) wireless communication systems are getting increased research attention because of their high capacity achieving capability. However, the practical implementation of MIMO systems rely on the computational complexity incurred in detection of the transmitted information symbols. The minimum bit error rate performance (BER) can be achieved by using maximum likelihood (ML) search based detection, but it is computationally impractical when number of transmit antennas increases. In this paper, we present a low-complexity hybrid algorithm (HA) to solve the symbol vector detection problem in large-MIMO systems. The proposed algorithm is inspired from the two well known bio-inspired optimization algorithms namely, particle swarm optimization (PSO) algorithm and ant colony optimization (ACO) algorithm. In the proposed algorithm, we devise a new probabilistic search approach which combines the distance based search of ants in ACO algorithm and the velocity based search of particles in PSO algorithm. The motivation behind using the hybrid of ACO and PSO is to avoid premature convergence to a local solution and to improve the convergence rate. Simulation results show that the proposed algorithm outperforms the popular minimum mean squared error (MMSE) algorithm and the existing ACO algorithms in terms of BER performance while achieve a near ML performance which makes the algorithm suitable for reliable detection in large-MIMO systems. Furthermore, a faster convergence to achieve a target BER is observed which results in reduction in computational efforts.     

呼叫中心系统中回声抵消算法的研究与实现

针对IP呼叫中心系统中的回声问题,对算法LMS（Least Mean Square）进行了研究。经研究发现,回声消除算法LMS在收敛速度与稳态误差之间始终存在着矛盾,即加快收敛速度,则稳态误差随之加大;减小稳态误差,则收敛速度随之减慢。研究的目的就是能够使两者之间的矛盾得到改善,能够更好地消除回声。研究的方法是在现有算法的基础上,提出了一种改进的LMS算法,通过计算机对该算法进行了仿真,以及利用DSP进行了回声消除,结论表明该算法具有良好的收敛性能和稳态性能,更好地改善了收敛速度与稳态误差之间的矛盾,消除效果较好。     

An improved algorithm for neural network classification ofimbalanced training sets

The backpropagation algorithm converges very slowly for two-class problems in which most of the exemplars belong to one dominant class. An analysis shows that this occurs because the computed net error gradient vector is dominated by the bigger class so much that the net error for the exemplars in the smaller class increases significantly in the initial iteration. The subsequent rate of convergence of the net error is very low. A modified technique for calculating a direction in weight-space which decreases the error for each class is presented. Using this algorithm, the rate of learning for two-class classification problems is accelerated by an order of magnitude.     

A weight initialization method for improving training speed in feedforward neural network

An algorithm for determining the optimal initial weights of feedforward neural networks based on the Cauchy's inequality and a linear algebraic method is developed. The algorithm is computational efficient. The proposed method ensures that the outputs of neurons are in the active region and increases the rate of convergence. With the optimal initial weights determined, the initial error is substantially smaller and the number of iterations required to achieve the error criterion is significantly reduced. Extensive tests were performed to compare the proposed algorithm with other algorithms. In the case of the sunspots prediction, the number of iterations required for the network initialized with the proposed method was only 3.03% of those started with the next best weight initialization algorithm.     

一种信号适度失真的宽线性SAEC模型

现有的立体声回声抵消器是一个实变量双输入双输出的装置,其结构复杂不易实现。宽线性模型的引入,提供了一种复变量单输入单输出的装置来替代实变量双输入双输出装置,其优点是只需处理一个复变量的输出信号而不是两个实变量输出信号,而且能通过复变量输入信号的相位和幅值分别调控声音的立体感和音质。利用输入信号适度失真的方法降低两个信号之间的相关性以解决因滤波而产生的非唯一性问题。把宽线性模型和失真信号应用到仿射投影算法中,通过仿真验证改进方法的误差性能和收敛速度。结果表明改进的方法具有误差小和收敛快的特点,因此宽线性SAEC模型更有优势。     

Time complexity of a path formulated optimal routing algorithm(second printing)

A detailed analysis of convergence rate is presented for an iterative path formulated optimal routing algorithm. In particular, it is quantified, analytically, how the convergence rate changes as the number of nodes in the underlying graph increases. The analysis is motivated by a particular path formulated gradient projection algorithm that has demonstrated excellent convergence rate properties through extensive numerical studies. The analytical result proven in this note is that the number of iterations for convergence depends on the number of nodes only through the network diameter     

Time complexity of a path formulated optimal routing algorithm

A detailed analysis of convergence rate is presented for an iterative path formulated optimal routing algorithm. In particular, it is quantified, analytically, how the convergence rate changes as the number of nodes in the underlying graph increases. The analysis is motivated by a particular path formulated gradient projection algorithm that has demonstrated excellent convergence rate properties through extensive numerical studies. The analytical result proven in this note is that the number of iterations for convergence depends on the number of nodes only through the network diameter     

一种优化WSNs节点部署的变步长虚拟力算法

针对传统虚拟力算法的后期稳定性较差,容易导致覆盖率降低的问题,提出了一种基于sigmoid函数的变步长虚拟力算法,通过每次迭代减小误差的方法,调整节点每次移动的步长,即节点移动的速度,提高收敛速度和后期稳定性.采用0—1圆盘节点感知模型,在800 m×700 m的矩形监测区域内,对提出的算法进行了仿真研究.仿真结果表明:与传统虚拟力算法相比,所提算法在保证收敛速度的同时,覆盖率均值提高了4.23%,覆盖率最优值提高了1.52%,稳定性提高了95.05%.     

正则系统在Lebesgue-p范数意义下的快速迭代学习控制

针对一类线性正则系统,传统迭代学习控制算法收敛速度较低的问题,设计了一种快速迭代学习控制算法。该算法在传统P型迭代学习控制算法基础上,增加了由相邻两次迭代时跟踪误差构成的上一次差分信号和当前差分信号,并在Lebesgue-p范数度量意义下,利用卷积推广的Young不等式严格证明了,当迭代次数趋于无穷大时,系统的跟踪误差收敛于零,并给出了算法的收敛条件。该算法与传统P型迭代学习控制算法相比,不仅提高了收敛速度,而且还避免了采用λ范数度量跟踪误差的缺陷,最后通过仿真结果进一步验证了所提算法的有效性。     

A local linearized least squares algorithm for training feedforwardneural networks

In training the weights of a feedforward neural network, it is well known that the global extended Kalman filter (GEKF) algorithm has much better performance than the popular gradient descent with error backpropagation in terms of convergence and quality of solution. However, the GEKF is very computationally intensive, which has led to the development of efficient algorithms such as the multiple extended Kalman algorithm (MEKA) and the decoupled extended Kalman filter algorithm (DEKF), that are based on dimensional reduction and/or partitioning of the global problem. In this paper we present a new training algorithm, called local linearized least squares (LLLS), that is based on viewing the local system identification subproblems at the neuron level as recursive linearized least squares problems. The objective function of the least squares problems for each neuron is the sum of the squares of the linearized backpropagated error signals. The new algorithm is shown to give better convergence results for three benchmark problems in comparison to MEKA, and in comparison to DEKF for highly coupled applications. The performance of the LLLS algorithm approaches that of the GEKF algorithm in the experiments.     

高斯混合交互式多模型容积信息滤波算法

针对机动目标跟踪中由于目标机动使系统的非线性强度增大,导致系统的线性误差增大和跟踪精度明显下降、甚至发散的问题,提出了基于高斯混合的交互式多模型容积信息滤波（ GMIMM-CIF）算法,实现对机动目标的精确跟踪。新算法在每次输入交互之后,保留概率较大的几个假设,并利用一个高斯混合项替换最优多模型算法中剩余的假设,从而使算法中假设的数量保持恒定;用容积信息滤波器（ CIF）代替传统的非线性滤波器,通过估计信息状态向量和信息矩阵而不是估计状态向量和协方差,可以减小系统的非线性误差。通过仿真对比实验,验证了该算法可以提高机动目标的跟踪精度。     

Performance analysis of under-modelling stereophonic acoustic echo cancellation by adaptive filtering LMS algorithm

This paper addresses the field of stereophonic acoustic echo cancellation (SAEC) with adaptive filtering algorithms. In SAEC applications, using the least mean square (LMS) algorithm, it is usually assumed that the lengths of the adaptive filters are equal to that of the unidentified system responses. Although, in many realistic situations, under-modelled lengths adaptive filters, whose lengths are less than that of the unidentified systems (under-modelled systems), are employed, and analysis results for the exact modelled stereophonic LMS algorithm are not automatically appropriate to the under-modeled lengths. In this paper, we present a statistical analysis of the under-modeled stereophonic LMS algorithm. Exact expressions and deterministic recursive equations to the mean coefficients behavior of the adaptive LMS filters are derived to completely characterize and assess the performances (transient and steady-state) of the under-modeling stereophonic LMS algorithm. The expected theoretical behaviour is compared with Monte Carlo simulations and practical experimental results, showing a very good agreement.     

快速收敛参考独立分量分析方法

为提高参考独立分量分析收敛速度,提出了改进的快速收敛参考独立分量分析方法。该方法首先采用预白化方法对观测到的信号进行处理,降低后续处理算法的复杂度,然后采用收敛速度更快、结构更简单的牛顿迭代方法对其进行优化,提高其收敛速度。理论分析表明,改进后的算法相对现有算法每次迭代的计算量基本相同;实验仿真结果表明,相对现有方法,改进后的算法具有较快的收敛速度,且误差保持不变。     

基于非线性Volterra信道的复数神经多项式盲均衡算法

针对传统常模算法收敛速度慢、均方误差大以及传统神经网络参数多、复杂度高的问题,提出了基于非线性Volterra信道的复数神经多项式盲均衡算法(Fuzzy neural network-complex valued neural polynomial-constant modulus algorithm,FNN -CNP-CMA）。该算法包含单层神经网络和非线性处理器的复数神经多项式,模块结构简单、复杂度低。由模糊神经网络(Fuzzy neural network, FNN）设计的模糊规则控制器能有效提高步长的控制精度。仿真实验结果表明,该算法系统结构简单、复杂度低、收敛速度快且稳态误差小,较好地解决了收敛速度与均方误差之间存在的矛盾。     

双蝙蝠群智能优化的多模盲均衡算法

针对常模盲均衡算法(CMA)均衡多模QAM信号收敛速度慢、剩余均方误差大的缺陷,提出了一种基于双蝙蝠群智能优化的多模盲均衡算法(DBSIO-MMA)。该算法将2个蝙蝠群独立全局寻优得到的一组最优位置向量分别作为多模盲均衡算法(MMA)初始化最优权向量的实部与虚部,以此提高收敛速度并减小剩余均方误差。仿真结果表明,蝙蝠算法(BA)全局搜索成功率高、收敛速度快的特点在DBSIO-MMA中得到很好地体现。与CMA、MMA、粒子群多模盲均衡算法(PSO-MMA)、单蝙蝠群多模盲均衡算法(BA-MMA)相比,DBSIO-MMA具有更快的收敛速度和更小的均方误差。