A new design method for the complex-valued multistate Hopfield associative memory

A method to store each element of an integral memory set M subset {1,2,...,K}/sup n/ as a fixed point into a complex-valued multistate Hopfield network is introduced. The method employs a set of inequalities to render each memory pattern as a strict local minimum of a quadratic energy landscape. Based on the solution of this system, it gives a recurrent network of n multistate neurons with complex and symmetric synaptic weights, which operates on the finite state space {1,2,...,K}/sup n/ to minimize this quadratic functional. Maximum number of integral vectors that can be embedded into the energy landscape of the network by this method is investigated by computer experiments. This paper also enlightens the performance of the proposed method in reconstructing noisy gray-scale images.     

Stability of fully asynchronous discrete-time discrete-state dynamic networks

We consider networks of a large number of neurons (or units, processors, ...), whose dynamics are fully asynchronous with overlapping updating. We suppose that the neurons take a finite number of states (discrete states), and that the updating scheme is discrete in time. We make no hypotheses on the activation function of the neurons; the networks may have multiple cycles and basins. We derive conditions on the initialization of the networks, which ensures convergence to fixed points only. Application to a fully asynchronous Hopfield neural network allows us to validate our study.     

An analog feedback associative memory

A method for the storage of analog vectors, i.e., vectors whose components are real-valued, is developed for the Hopfield continuous-time network. An important requirement is that each memory vector has to be an asymptotically stable (i.e. attractive) equilibrium of the network. Some of the limitations imposed by the continuous Hopfield model on the set of vectors that can be stored are pointed out. These limitations can be relieved by choosing a network containing visible as well as hidden units. An architecture consisting of several hidden layers and a visible layer, connected in a circular fashion, is considered. It is proved that the two-layer case is guaranteed to store any number of given analog vectors provided their number does not exceed 1 + the number of neurons in the hidden layer. A learning algorithm that correctly adjusts the locations of the equilibria and guarantees their asymptotic stability is developed. Simulation results confirm the effectiveness of the approach.     

A Boolean Hebb rule for binary associative memory design

A binary associative memory design procedure that gives a Hopfield network with a symmetric binary weight matrix is introduced in this paper. The proposed method is based on introducing the memory vectors as maximal independent sets to an undirected graph, which is constructed by Boolean operations analogous to the conventional Hebb rule. The parameters of the resulting network is then determined via the adjacency matrix of this graph in order to rind a maximal independent set whose characteristic vector is close to the given distorted vector. We show that the method provides attractiveness for each memory vector and avoids spurious memories whenever the set of given memory vectors satisfy certain compatibility conditions, which implicitly imply sparsity. The applicability of the design method is finally investigated by a quantitative analysis of the compatibility conditions.     

Continuous‐ and discrete‐time fixed‐gain controller designs for the control of vehicle lateral dynamics

In this paper, fixed‐gain feedback linearization controls are presented to stabilize the vehicle lateral dynamics at bifurcation points for both continuous‐time and discrete‐time cases. Based on the assumption of constant driving speed, a second‐order nonlinear lateral dynamics model is adopted for controller design. Via the feedback linearization scheme and the first‐order Taylor series expansion, a time‐invariant feedback linearization control is proposed as a fixed‐gain linear version of the previously proposed nonlinear one. Furthermore, the conventional linear quadratic regulator (LQR) design is applied to facilitate the choice of the fixed‐gain matrix. Refined controls to compensate the model uncertainty and their local stability analysis are provided. Extension of the continuous‐time design results to discrete‐time cases is also addressed. Numerical simulations for an example model demonstrate the effectiveness of the proposed continuous‐time and discrete‐time design results. Copyright © 2010 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society     

Associative Memory Design via Path Embedding into a Graph

A graph theoretical procedure for storing a set of n-dimensional binary vectors as asymptotically stable equilibrium points of a discrete Hopfield neural network is presented. The method gives an auto-associative memory which stores an arbitrary memory set completely. Spurious memories might occur only in a small neighborhood of the original memory vectors, so cause small errors.     

A theoretical investigation into the performance of the Hopfieldmodel

An analysis is made of the behavior of the Hopfield model as a content-addressable memory (CAM) and as a method of solving the traveling salesman problem (TSP). The analysis is based on the geometry of the subspace set up by the degenerate eigenvalues of the connection matrix. The dynamic equation is shown to be equivalent to a projection of the input vector onto this subspace. In the case of content-addressable memory, it is shown that spurious fixed points can occur at any corner of the hypercube that is on or near the subspace spanned by the memory vectors. Analysed is why the network can frequently converge to an invalid solution when applied to the traveling salesman problem energy function. With these expressions, the network can be made robust and can reliably solve the traveling salesman problem with tour sizes of 50 cities or more.     

用改进的差分式Hopfield网络实现线性二次型最优控制

为解决差分式Hopfield网络能量函数的局部极小问题,本文对之改进得到一种具有迭代学习功能的线性差分式Hopfield网络.理论分析表明,该网络具有稳定性,且稳定状态使其能量函数达到唯一极小值.基于线性差分式Hopfield网络稳定性与其能量函数收敛特性的关系,本文将该网络用于求解多变量时变系统的线性二次型最优控制问题.网络的理论设计方法表明,网络的稳态输出就是欲求的最优控制向量.数字仿真取得了与理论分析一致的实验结果.     

Discrete plane segmentation and estimation from a point cloud using local geometric patterns

This paper presents a method for segmenting a 3D point cloud into planar surfaces using recently obtained discretegeometry results. In discrete geometry, a discrete plane is defined as a set of grid points lying between two parallel planes with a small distance, called thickness. In contrast to the continuous case, there exist a finite number of local geometric patterns （LGPs） appearing on discrete planes. Moreover, such an LGP does not possess the unique normal vector but a set of normal vectors. By using those LGP properties, we first reject non-linear points from a point cloud, and then classify non-rejected points whose LGPs have common normal vectors into a planar-surface-point set. From each segmented point set, we also estimate the values of parameters of a discrete plane by minimizing its thickness.     

一种基于离散马尔科夫模型的神经网络可靠性设计方法

本文给出了一和循环前馈神经网络来表示具有ｎ个状态的离莠马尔科夫模型,及其二次能量函数,并根据梯度下降规则,给出了相应的权值修改规则。利用该神经网络方法可以有效地对容错硬件进行可靠性设计。     

Backwards type analysis of asynchronous method calls

Asynchronous method calls have been proposed to better integrate object orientation with distribution. In the language, asynchronous method calls are combined with so-called processor release points in order to allow concurrent objects to adapt local scheduling to network delays in a very flexible way. However, asynchronous method calls complicate the type analysis by decoupling input and output information for method calls, which can be tracked by a type and effect system. Interestingly, backwards type analysis simplifies the effect system considerably and allows analysis in a single pass. This paper presents a kernel language with asynchronous method calls and processor release points, a novel mechanism for local memory deallocation related to asynchronous method calls, an operational semantics in rewriting logic for the language, and a type and effect system for backwards analysis. Source code is translated into runtime code as an effect of the type analysis, automatically inserting inferred type information in method invocations and operations for local memory deallocation in the process. We establish a subject reduction property, showing in particular that method lookup errors do not occur at runtime and that the inserted deallocation operations are safe.     

Basins of attraction in fully asynchronous discrete-time discrete-state dynamic networks

This paper gives a formulation of the basins of fixed point states of fully asynchronous discrete-time discrete-state dynamic networks. That formulation provides two advantages. The first one is to point out the different behaviors between synchronous and asynchronous modes and the second one is to allow us to easily deduce an algorithm which determines the behavior of a network for a given initialization. In the context of this study, we consider networks of a large number of neurons (or units, processors, etc.), whose dynamic is fully asynchronous with overlapping updates . We suppose that the neurons take a finite number of discrete states and that the updating scheme is discrete in time. We make no hypothesis on the activation functions of the nodes, so that the dynamic of the network may have multiple cycles and/or basins. Our results are illustrated on a simple example of a fully asynchronous Hopfield neural network.     

Optimal twist vectors as a tool for interpolating a network of curves with a minimum energy surface

This paper provides a new answer to the old problem of specifying the mixed partial derivatives (MPDs) or ‘twist vectors’ at the grid points for an interpolating surface over a rectangular network of curves. An algorithm is presented for finding the MPDs that minimizes a generalized energy integral over the entire surface. The integrand may be any quadratic form in the second partial derivatives of the surface. This results in a surface design technique for interpolating over a network of curves by automatically selecting the optimal twist vectors at the grid points.     

连续时间 Hopfield网络模型数值实现分析

讨论使用Euler方法和梯形方法在数值求解连续时间的Hopfield网络模型时,离散时间步长的选择和迭代停止条件问题.利用凸函数的定义研究了能量函数下降的条件,根据凸函数的性质分析它的共轭函数减去二次函数之差仍为凸函数的条件.分析连续时间Hopfield网络模型的收敛性证明,提出了一个广义的连续时间Hopfield网络模型.对于常用的Euler方法和梯形方法数值求数值实现连续时间Hopfield网络,讨论了离散时间步长的选择.由于梯形方法为隐式方法,分析了它的迭代求算法的停止条件.根据连续时间Hopfield网络的特点,提出改进的迭代算法,并对其进行了分析.数值实验的结果表明,较大的离散时间步长不仅加速了数值实现,而且有利于提高优化性能.     

余弦调制子带滤波器组迭代设计的改进算法

余弦调制子带滤波器组的设计可以转化成一个高维非线性优化问题,传统迭代设计方法虽然避免了复杂的直接求解过程,但是其混叠误差的抑制依赖于滤波器阻带衰减的大小,设计出的滤波器往往阶数过高。本文提出了一种改进的子带余弦调制滤波器组迭代设计算法,其代价函数除了包括子带系统响应和子带滤波器阻带能量之外,还添加了系统一次混叠项因子,因此更好地抑制了混叠误差。本文将新的代价函数表示成滤波器系数的二次函数形式,并推导出其极值点的闭合解表达式。算法保留了传统迭代方法逐渐逼近极值点的思想,简单易于实现。仿真结果表明与传统迭代设计方法相比,本文算法获得的子带滤波器组具有更小的混叠误差和更低的滤波器阶数。     

A study of the fixed points and spurious solutions of the deflation-based FastICA algorithm

The FastICA algorithm is one of the most popular algorithms in the domain of independent component analysis (ICA). Despite its success, it is observed that FastICA occasionally yields outcomes that do not correspond to any true solutions (known as demixing vectors) of the ICA problem. These outcomes are commonly referred to as spurious solutions. Although FastICA is a well-studied ICA algorithm, the occurrence of spurious solutions is not yet completely understood by the ICA community. In this contribution, we aim at addressing this issue. In the first part of this work, we are interested in characterizing the relationship between demixing vectors, local optimizers of the contrast function and (attractive or unattractive) fixed points of FastICA algorithm. We will show that there exists an inclusion relationship between these sets. In the second part, we investigate the possible scenarios where spurious solutions occur. It will be shown that when certain bimodal Gaussian mixture distributions are involved, there may exist spurious solutions that are attractive fixed points of FastICA. In this case, popular nonlinearities such as “Gauss” or “tanh” tend to yield spurious solutions, whereas “kurtosis” gives much more reliable results.

     

Local function approximation in evolutionary algorithms for the optimization of costly functions

We develop an approach for the optimization of continuous costly functions that uses a space-filling experimental design and local function approximation to reduce the number of function evaluations in an evolutionary algorithm. Our approach is to estimate the objective function value of an offspring by fitting a function approximation model over the k nearest previously evaluated points, where k=(d+1)(d+2)/2 and d is the dimension of the problem. The estimated function values are used to screen offspring to identify the most promising ones for function evaluation. To fit function approximation models, a symmetric Latin hypercube design (SLHD) is used to determine initial points for function evaluation. We compared the performance of an evolution strategy (ES) with local quadratic approximation, an ES with local cubic radial basis function (RBF) interpolation, an ES whose initial parent population comes from an SLHD, and a conventional ES. These algorithms were applied to a twelve-dimensional (12-D) groundwater bioremediation problem involving a complex nonlinear finite-element simulation model. The performances of these algorithms were also compared on the Dixon-Szego test functions and on the ten-dimensional (10-D) Rastrigin and Ackley test functions. All comparisons involve analysis of variance (ANOVA) and the computation of simultaneous confidence intervals. The results indicate that ES algorithms with local approximation were significantly better than conventional ES algorithms and ES algorithms initialized by SLHDs on all Dixon-Szego test functions except for Goldstein-Price. However, for the more difficult 10-D and 12-D functions, only the cubic RBF approach was successful in improving the performance of an ES. Moreover, the results also suggest that the cubic RBF approach is superior to the quadratic approximation approach on all test functions and the difference in performance is statistically significant for all test functions with dimension d/spl ges/4.     

Linux系统下实现网络内存共享关键技术探析

网络内存共享的出现主要借鉴了传统的网格计算技术和集群内存共享技术。论文从动态函数截获,缓存数据组织与管理和异步缓存数据写入三方面探讨了网络内存共享的关键技术。     

A two-level approach to implicit surface modeling with compactly supported radial basis functions

We describe a two-level method for computing a function whose zero-level set is the surface reconstructed from given points scattered over the surface and associated with surface normal vectors. The function is defined as a linear combination of compactly supported radial basis functions (CSRBFs). The method preserves the simplicity and efficiency of implicit surface interpolation with CSRBFs and the reconstructed implicit surface owns the attributes, which are previously only associated with globally supported or globally regularized radial basis functions, such as exhibiting less extra zero-level sets, suitable for inside and outside tests. First, in the coarse scale approximation, we choose basis function centers on a grid that covers the enlarged bounding box of the given point set and compute their signed distances to the underlying surface using local quadratic approximations of the nearest surface points. Then a fitting to the residual errors on the surface points and additional off-surface points is performed with fine scale basis functions. The final function is the sum of the two intermediate functions and is a good approximation of the signed distance field to the surface in the bounding box. Examples of surface reconstruction and set operations between shapes are provided.     

Minimum-seeking properties of analog neural networks withmultilinear objective functions

In this paper, we study the problem of minimizing a multilinear objective function over the discrete set {0, 1}ⁿ. This is an extension of an earlier work addressed to the problem of minimizing a quadratic function over {0, 1}ⁿ. A gradient-type neural network is proposed to perform the optimization. A novel feature of the network is the introduction of a so-called bias vector. The network is operated in the high-gain region of the sigmoidal nonlinearities. The following comprehensive theorem is proved: For all sufficiently small bias vectors except those belonging to a set of measure zero, for all sufficiently large sigmoidal gains, for all initial conditions except those belonging to a nowhere dense set, the state of the network converges to a local minimum of the objective function. This is a considerable generalization of earlier results for quadratic objective functions. Moreover, the proofs here are completely rigorous. The neural network-based approach to optimization is briefly compared to the so-called interior-point methods of nonlinear programming, as exemplified by Karmarkar's algorithm. Some problems for future research are suggested