Memetic evolutionary training for recurrent neural networks: an application to time-series prediction

Abstract: Artificial neural networks are bio-inspired mathematical models that have been widely used to solve complex problems. The training of a neural network is an important issue to deal with, since traditional gradient-based algorithms become easily trapped in local optimal solutions, therefore increasing the time taken in the experimental step. This problem is greater in recurrent neural networks, where the gradient propagation across the recurrence makes the training difficult for long-term dependences. On the other hand, evolutionary algorithms are search and optimization techniques which have been proved to solve many problems effectively. In the case of recurrent neural networks, the training using evolutionary algorithms has provided promising results. In this work, we propose two hybrid evolutionary algorithms as an alternative to improve the training of dynamic recurrent neural networks. The experimental section makes a comparative study of the algorithms proposed, to train Elman recurrent neural networks in time-series prediction problems.     

Evolving neurocontrollers for balancing an inverted pendulum

This paper introduces an evolutionary algorithm that is tailored to generate recurrent neural networks functioning as nonlinear controllers. Network size and architecture, as well as network parameters like weights and bias terms, are developed simultaneously. There is no quantization of inputs, outputs or internal parameters. Different kinds of evolved networks are presented that solve the pole-balancing problem, i.e. balancing an inverted pendulum. In particular, controllers solving the problem for reduced phase space information (only angle and cart position) use a recurrent connectivity structure. Evolved controllers of 'minimal' size still have a very good benchmark performance.     

Neural network design based on evolutionary programming

In this paper, a new design method for neural networks is presented based on evolutionary programming. By using an evolutionary algorithm the structure and weights of static neural networks can be simultaneously acquired. This method is further extended to design recurrent neural networks through introducing ‘delayed links’ into networks. Simulation results are also given to illustrate the efficiency of the proposed method.     

Multi-objective evolutionary algorithm using problem-specific genetic operators for community detection in networks

Automatic network clustering is an important method for mining the meaningful communities of complex networks. Uncovered communities help to understand the potential system structure and functionality. Many algorithms that use multiple optimization criteria and optimize a population of solutions are difficult to apply to real systems because they suffer a long optimization process. In this paper, in order to accelerate the optimization process and to uncover multiple significant community structures more effectively, a multi-objective evolutionary algorithm is proposed and evaluated using problem-specific genetic mutation and group crossover, and problem-specific initialization. Since crossover operators mainly contribute to performance of genetic algorithms, more problem-specific group crossover operators are introduced and evaluated for intelligent evolution of population. The experiments on both artificial and real-world networks demonstrate that the proposed evolutionary algorithm with problem-specific genetic operations has effective performance on discovering the community structure of networks.

     

进化算法在复杂网络中的应用综述

复杂网络作为现今科学研究中的一个热点学科,在过去20年里得到了巨大的发展.现实中大量的复杂的交互系统,比如互联网、交通运输网、神经网络等都可以抽象为复杂网络,以进行系统的分析和研究.进化算法作为优化工具应用于复杂网络的不同领域的各个任务中,如网络社团结构的检测任务、网络动力学中的鲁棒性优化任务、网络传播中关键节点的搜寻任务等.本文首先对复杂网络和进化算法相关的基础知识进行了全面的概述,重点讨论了复杂网络中目标优化的研究进展,针对不同任务对优化目标及其具体应用展开了详细介绍,同时,对算法的性能评价指标进行了概述.此外,本文通过一系列实验展示了单/多目标优化算法在复杂网络优化问题上的性能表现,以及部分目标之间的相关性关系.最后对复杂网络中优化问题未来的研究动向进行了展望,为今后研究人员开展进化计算和复杂网络相结合的相关研究提供一些思路.     

基于实数编码遗传算法的神经网络优化设计

提出一种基于综合控制策略的改进的实数编码遗传算法,用该算法对前向神经网络的结构及权值进行优化。通过实验结果表明,该算法能快速有效的确定网络的结构及权值。     

实数编码遗传算法的前向神经网络优化设计

提出一种综合控制策略的实数编码遗传算法,用该算法实现对前向网络结构及权值的同时优化设计。用非线性函数的逼近问题作仿真实验。结果表明,该算法能快速有效地确定网络结构及权值。     

链路预测的方法与发展综述

链路预测对网络结构特征的演化趋势进行挖掘有着不可磨灭的促进作用。为了对网络的未来结构变化进行预测，学者们提出了许多算法。综述了4类较为常见的链路预测方法，分别是基于节点属性、基于网络拓扑结构、基于机器学习以及基于最大似然的方法，比较了4类预测方法的优劣，并概述了几种常见的衡量链路预测算法精确度标准。最后总结并展望了链路预测的未来研究方向和发展前景。     

改进的遗传算法在优化BP网络权值中的应用

对遗传算法和BP神经网络的特点进行了比较,作为进化算法神经网络与遗传算法的目标相近而方法各异。阐述了遗传算法与神经网络结合的必要性。提出了一种改进的遗传算法优化BP神经网络的权值,用遗传算法的全局随机搜索能力弥补了神经网络容易陷入局部最优解的问题。同时,在遗传算法中改变传统的同代交叉机制,采用父代与子代进行交叉,避免了遗传算法过早丧失进化能力。     

Advances in evolutionary feature selection neural networks with co-evolution learning

Training neural networks is a complex task provided that many algorithms are combined to find best solutions to the classification problem. In this work, we point out the evolutionary computing to minimize a neural configuration. For this purpose, a distribution estimation framework is performed to select relevant features, which lead to classification accuracy with a lower complexity in computational time. Primarily, a pruning strategy-based score function is applied to decide the network relevance in the genetic population. Since the complexity of the network (connections, weights, and biases) is most important, the cooling state of the system will strongly relate to the entropy as a minimization function to reach the desired solution. Also, the framework proposes coevolution learning (with discrete and continuous representations) to improve the behavior of the evolutionary neural learning. The results obtained after simulations show that the proposed work is a promising way to extend its usability to other classes of neural networks.     

Neurocontrol of nonlinear dynamical systems with Kalman filtertrained recurrent networks

Although the potential of the powerful mapping and representational capabilities of recurrent network architectures is generally recognized by the neural network research community, recurrent neural networks have not been widely used for the control of nonlinear dynamical systems, possibly due to the relative ineffectiveness of simple gradient descent training algorithms. Developments in the use of parameter-based extended Kalman filter algorithms for training recurrent networks may provide a mechanism by which these architectures will prove to be of practical value. This paper presents a decoupled extended Kalman filter (DEKF) algorithm for training of recurrent networks with special emphasis on application to control problems. We demonstrate in simulation the application of the DEKF algorithm to a series of example control problems ranging from the well-known cart-pole and bioreactor benchmark problems to an automotive subsystem, engine idle speed control. These simulations suggest that recurrent controller networks trained by Kalman filter methods can combine the traditional features of state-space controllers and observers in a homogeneous architecture for nonlinear dynamical systems, while simultaneously exhibiting less sensitivity than do purely feedforward controller networks to changes in plant parameters and measurement noise.     

An approach to reservoir computing design and training

Reservoir computing is a framework for computation like a recurrent neural network that allows for the black box modeling of dynamical systems. In contrast to other recurrent neural network approaches, reservoir computing does not train the input and internal weights of the network, only the readout is trained. However it is necessary to adjust parameters to create a “good” reservoir for a given application. In this study we introduce a method, called RCDESIGN (reservoir computing and design training). RCDESIGN combines an evolutionary algorithm with reservoir computing and simultaneously looks for the best values of parameters, topology and weight matrices without rescaling the reservoir matrix by the spectral radius. The idea of adjust the spectral radius within the unit circle in the complex plane comes from the linear system theory. However, this argument does not necessarily apply to nonlinear systems, which is the case of reservoir computing. The results obtained with the proposed method are compared with results obtained by a genetic algorithm search for global parameters generation of reservoir computing. Four time series were used to validate RCDESIGN.     

A multiobjective genetic algorithm for obtaining the optimal size of a recurrent neural network for grammatical inference

Grammatical inference has been extensively studied in recent years as a result of its wide field of application, and in turn, recurrent neural networks have proved themselves to be a good tool for grammatical inference. The learning algorithms for these neural networks, however, have been far less studied than those for feed-forward neural networks. Classical training methods for recurrent neural networks suffer from being trapped in local minimal and having a high computational time. In addition, selecting the optimal size of a neural network for a particular application is a difficult task. This suggests that the problems of developing methods to determine optimal topologies and new training algorithms should be studied.In this paper, we present a multi-objective evolutionary algorithm which is able to determine the optimal size of recurrent neural networks in any particular application. This is specially analyzed in the case of grammatical inference: in particular, we study how to establish the optimal size of a recurrent neural network in order to learn positive and negative examples in a certain language, and how to determine the corresponding automaton using a self-organizing map once the training has been completed.     

Recurrent neural networks and prior knowledge for sequence processing: a constrained nondeterministic approach

The paper focuses on methods for injecting prior knowledge into adaptive recurrent networks for sequence processing. In order to increase the flexibility needed for specifying partially known rules, a nondeterministic approach for modelling domain knowledge is proposed. The algorithms presented in the paper allow time-warping nondeterministic automata to be mapped into recurrent architectures with first-order connections. These kinds of automata are suitable for modeling temporal scale distortions in data such as acoustic sequences occurring in problems of speech recognition. The algorithms output a recurrent architecture and a feasible region in the connection weight space. It is demonstrated that, as long as the weights are constrained into the feasible region, the nondeterministic rules introduced using prior knowledge are not destroyed by learning. The paper focuses primarily on architectural issues, but the proposed method allows the connection weights to be subsequently tuned to adapt the behavior of the network to data.     

Multi-level learning based memetic algorithm for community detection

Complex network has become an important way to analyze the massive disordered information of complex systems, and its community structure property is indispensable to discover the potential functionality of these systems. The research on uncovering the community structure of networks has attracted great attentions from various fields in recent years. Many community detection approaches have been proposed based on the modularity optimization. Among them, the algorithms which optimize one initial solution to a better one are easy to get into local optima. Moreover, the algorithms which are susceptible to the optimized order are easy to obtain unstable solutions. In addition, the algorithms which simultaneously optimize a population of solutions have high computational complexity, and thus they are difficult to apply to practical problems. To solve the above problems, in this study, we propose a fast memetic algorithm with multi-level learning strategies for community detection by optimizing modularity. The proposed algorithm adopts genetic algorithm to optimize a population of solutions and uses the proposed multi-level learning strategies to accelerate the optimization process. The multi-level learning strategies are devised based on the potential knowledge of the node, community and partition structures of networks, and they work on the network at nodes, communities and network partitions levels, respectively. Extensive experiments on both benchmarks and real-world networks demonstrate that compared with the state-of-the-art community detection algorithms, the proposed algorithm has effective performance on discovering the community structure of networks.     

Dual modularity optimization for detecting overlapping communities in bipartite networks

Many algorithms have been designed to discover community structure in networks. These algorithms are mostly dedicated to detecting disjoint communities. Very few of them are intended to discover overlapping communities, particularly the bipartite networks have hardly been explored for the detection of such communities. In this paper, we describe a new approach which consists in forming overlapping mixed communities in a bipartite network based on dual optimization of modularity. To this end, we propose two algorithms. The first one is an evolutionary algorithm dedicated for global optimization of the Newman’s modularity on the line graph. This algorithm has been tested on well-known real benchmark networks and compared with several other existing methods of community detection in networks. The second one is an algorithm that locally optimizes the graph Mancoridis modularity, and we have adapted to a bipartite graph. Specifically, this second algorithm is applied to the decomposition of vertices, resulting from the evolutionary process, and also characterizes the overlapping communities taking into account their semantic aspect. Our approach requires a priori no knowledge on the number of communities searched in the network. We show its interest on two datasets, namely, a group of synthetic networks and real-world network whose structure is also difficult to understand.     

基于郭涛算法的演化神经网络

提出了一种新的演化神经网络算法GTEANN,该算法基于高效的郭涛算法,同时完成在网络结构空间和权值空间的搜索,以实现前馈神经网络的自动化设计。本方法采用的编码方案直观有效,基于该编码表示,神经网络的学习过程是一个复杂的混合整实数非线性规划问题,例如杂交操作包括网络的同构和规整处理。初步实验结果表明该方法收敛,能够达到根据训练样本自动优化设计多层前馈神经网络的目的。     

复杂网络重叠社区结构发现的演化算法研究

复杂网络重叠社区结构的划分已成为复杂网络研究的一个热点,目前已提出了很多关于社区结构发现的算法。提出了一种基于个体从众的演化算法ICEA,基本思想是由节点邻居组成的个体依概率进行从众和变异操作,用较短时间找到最优(或拟最优)模块度的社区划分,社区结构确定后利用邻居投票机制NV发现网络的重叠节点,完成重叠社区的划分。在真实网络的实验结果表明,此算法的使用时间和划分结果都优于典型算法。