Self-organizing Learning Algorithm Research Articles

A Trade-Off Task-Offloading Scheme in Multi-User Multi-Task Mobile Edge Computing

Mobile edge computing is a novel technique that can reduce computational burden of local terminal devices by tasks offloading, which emerges as a promising paradigm to provide computing capabilities near mobile users. In this paper, the task offloading decision to solve the queuing problem of tasks to be processed in the local terminals was being considered, with an application of Lyapunov theory to ensure the queue stability. Then, a trade-off model was formulated to minimize the delay and energy consumption to achieve a minimum execution cost. Moreover, an improved Dynamic Niche-based Self-organizing Learning Algorithm was presented to accelerate the speed of the search process to gain an optimal task offloading scheme. The simulation results provided supporting evidence that the proposed optimal scheme IDO could achieve a lower average energy consumption than LFO and a lower average execution delay than EFO under the scenarios of the waiting queue length varies within [10, 50], the number of users varies within [4, 20], and the number of tasks varies within [20, 100].

A Hybrid Method Based on Extreme Learning Machine and Self Organizing Map for Pattern Classification.

Extreme learning machine is a fast learning algorithm for single hidden layer feedforward neural network. However, an improper number of hidden neurons and random parameters have a great effect on the performance of the extreme learning machine. In order to select a suitable number of hidden neurons, this paper proposes a novel hybrid learning based on a two-step process. First, the parameters of hidden layer are adjusted by a self-organized learning algorithm. Next, the weights matrix of the output layer is determined using the Moore–Penrose inverse method. Nine classification datasets are considered to demonstrate the efficiency of the proposed approach compared with original extreme learning machine, Tikhonov regularization optimally pruned extreme learning machine, and backpropagation algorithms. The results show that the proposed method is fast and produces better accuracy and generalization performances.

Vision-Based Robot Navigation through Combining Unsupervised Learning and Hierarchical Reinforcement Learning.

Extensive studies have shown that many animals’ capability of forming spatial representations for self-localization, path planning, and navigation relies on the functionalities of place and head-direction (HD) cells in the hippocampus. Although there are numerous hippocampal modeling approaches, only a few span the wide functionalities ranging from processing raw sensory signals to planning and action generation. This paper presents a vision-based navigation system that involves generating place and HD cells through learning from visual images, building topological maps based on learned cell representations and performing navigation using hierarchical reinforcement learning. First, place and HD cells are trained from sequences of visual stimuli in an unsupervised learning fashion. A modified Slow Feature Analysis (SFA) algorithm is proposed to learn different cell types in an intentional way by restricting their learning to separate phases of the spatial exploration. Then, to extract the encoded metric information from these unsupervised learning representations, a self-organized learning algorithm is adopted to learn over the emerged cell activities and to generate topological maps that reveal the topology of the environment and information about a robot’s head direction, respectively. This enables the robot to perform self-localization and orientation detection based on the generated maps. Finally, goal-directed navigation is performed using reinforcement learning in continuous state spaces which are represented by the population activities of place cells. In particular, considering that the topological map provides a natural hierarchical representation of the environment, hierarchical reinforcement learning (HRL) is used to exploit this hierarchy to accelerate learning. The HRL works on different spatial scales, where a high-level policy learns to select subgoals and a low-level policy learns over primitive actions to specialize on the selected subgoals. Experimental results demonstrate that our system is able to navigate a robot to the desired position effectively, and the HRL shows a much better learning performance than the standard RL in solving our navigation tasks.

Continuous learning route map for robot navigation using a growing-on-demand self-organizing neural network

This article proposes an experience-based route map continuous learning method and applies it into robot planning and navigation. First of all, the framework for robot route map learning and navigation is designed, which incorporates the four cyclic processes of planning, motion, perception, and extraction, enabling robot to constantly learn the information of the road experience and to obtain and improve the route map of the environment. Besides, a growing-on-demand self-organizing neural network learning algorithm is also proposed. This algorithm is based on growing neural gas algorithm, but it does not require presetting of network scale, and under the condition of dynamically growing input data, it can regulate the increase scale of network online in a self-adaptive and self-organized manner to obtain stable learning results. Finally, with robot roaming in an environment, this algorithm is used to conduct continuous learning of dynamically increasing route information, extract the topological structur...

Dynamic Niche-Based Self-Organizing Learning Algorithm

PDF HTML阅读 XML下载 导出引用 引用提醒 一种基于动态小生境的自组织学习算法 DOI: 10.3724/SP.J.1001.2011.03830 作者: 作者单位: 作者简介: 通讯作者: 中图分类号: 基金项目: 安徽省教育厅重大项目基金(ZD200904) Dynamic Niche-Based Self-Organizing Learning Algorithm Author: Affiliation: Fund Project: 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:提出了一种基于动态小生境的自组织学习算法(dynamic niche-based self-organizing learning algorithm,简称DNSLA),实现了基于0-1 编码的动态学习机制.种群中的个体由被动适应转为主动学习,即通过系统的自组织学习而实现与环境的友好交互,因而具有更强健的动态环境适应能力,能够及时、准确地侦测到环境的变化并跟踪极值点在搜索空间内的运动轨迹,具有良好的可移植性和很强的泛化能力.一系列动态测试问题的对比仿真实验结果表明,该算法即使在剧烈动荡的环境中也能很好地与环境进行稳定而友好的交互学习,表现出了很强的鲁棒性,其动态搜索能力和极值点跟踪能力远优于同类搜索方法. Abstract:A dynamic niche-based self-organizing learning algorithm (DNSLA) was proposed in this paper. The dynamic learning mechanism based on 0-1 coding method was carried out, and the individuals involved in this algorithm were able to adapt to the dynamic environments through active learning, which was different from the passive adaptive search strategy in traditional evolutionary algorithms. As a result of self-organizing learning and friendly interaction with the environments, DNSLA was more robust to adapt to the dynamic problems, and it was able to accurately detect the slight changes of the environments and track the extreme points in the solution domain. A series of dynamic simulation tests for comparative experiments showed that, even in the turbulent environments, DNSLA was still able to perform friendly interactive learning with the dynamic environments. DNSLA showed a strong robustness in the comparative experiments, whose dynamic search capabilities were far superior to other search methods. 参考文献 相似文献 引证文献

Behavioral Modeling of Power Amplifiers With Dynamic Fuzzy Neural Networks

In this letter, dynamic fuzzy neural networks (D-FNN) are applied to model power amplifiers (PAs) with memory effects. The D-FNN model implements Takagi-Sugeno-Kang (TSK) fuzzy systems based on extended radial bias function (RBF) neural networks. The parameters of the model are trained by the online self-organized learning algorithm, in which the neurons can be recruited or deleted dynamically according to their significance to system performance, and the over fitting or over training problems can be avoided. The D-FNN model is validated in our test bench in which a Doherty PA is excited with 10 MHz and 20 MHz worldwide interoperability for microwave access (WiMAX) signals. Experimental results show that the D-FNN model can give an accurate approximation to characterize the wideband PAs with memory effects.

Supervisory Data Alignment for Text-Independent Voice Conversion

We propose new supervisory data alignment methods for text-independent voice conversion which do not need parallel training corpora. Phonetic information is used as a restriction during alignment for mapping the data from the source speaker onto the parameter space of a target speaker. Both linear and nonlinear methods are derived by considering alignment accuracy and topology preservation. For the linear alignment, we consider common phoneme clusters of the source and target space as benchmarks and adapt the source data vector to the target space while maintaining the relative phonetic positions among neighborhood clusters. In order to preserve the topological structure of the source parameter space and improve the stability of conversion and the accuracy of the phonetic mapping, a supervised self-organizing learning algorithm considering phonetic restriction is proposed for iteratively improving the alignment outcome of the previous step. Both the linear and nonlinear methods can also be applied in the cross-lingual case. Evaluation results show that the proposed methods improve the performance of alignment in terms of both alignment accuracy and stability for text-independent voice conversion in intra-lingual and cross-lingual cases.

良否を含む顎矯正手術事例の自己組織化学習による集約とその手術計画立案への援用

本論文では，良否を含む過去の顎矯正手術事例から得られる知見を集約し，新規患者の手術計画に援用する手法を提案した．顎矯正手術において，審美性という点のみに注目して過去の症例を考察した場合，必ずしも審美性が向上したとは言えない事例が存在する．適切な手術計画を行うためには，審美的な観点から望ましい事例のみならず，望ましくない事例から得られる知見をも積極的に取り扱うことが必要である．本論文では，成否両方の事例から望ましい入出力関係の獲得を行うニューラルネットワークである自己組織化関係ネットワークを改良した手法を提案した．さらに，集約した知見を視覚的に理解しやすい形式で表現するために，画像処理による症例のクラスタリング手法などを導入した．提案手法とプロフィログラムなどを用いた従来手法とを統合した手術計画援用システムを構築し，その定性的評価を行った．

Dynamic recurrent radial basis function network model predictive control of unstable nonlinear processes

A multistep model predictive control (MPC) strategy based on dynamically recurrent radial basis function networks (RBFNs) is proposed for single-input single-output (SISO) control of uncertain nonlinear processes. The control system consists of two automatically configured RBFNs, a trained network representing the plant model and a network with on-line learning to function as controller. The automatic configuration and learning of the networks is carried out by using a hierarchically self-organizing learning algorithm. This control strategy is structurally simple and computationally efficient since a single output node of each RBFN is configured to provide multistep predictions for plant output and controller. The performance of the proposed RBFNMPC strategy is evaluated by applying to two unstable nonlinear chemical processes, a chemical reactor and a biochemical reactor, and also a stable polymerization reactor. Further, the results of the RBFNMPC is compared with similar RBFN model based control strategies and also with well tuned PID/PI controller. The results show the better performance of the proposed RBFNMPC for the control of open-loop unstable nonlinear processes that exhibit multiple steady-state behavior.

Modeling Connectionist Neuro-Fuzzy network and Applications

The main objective of this paper is to propose a Neuro-Fuzzy network, which can model a system from input–output data by automatically dividing the input–output space and extracting fuzzy if-then rules from numerical data. The structure of the network is simple with input, rule and output layers only. The connections between input and rule layer is used to derive the membership functions of the fuzzified part. Kohonen’s self-organizing learning algorithm is applied to partition the pattern space. Using this algorithm, similar rules are mapped close by and extraction of if-then rules is made easy. It can also adapt to a number of rules automatically. The proposed network is verified for three benchmark applications. Experimental results show that the adaptive method discussed in this paper not only trains in a few hundred iterations but also provides better performance measures when compared to conventional methods.

Self-Organizing Learning in Robotics

One of the most important features of biological motion control is the connection of motor action to sensory perception. Sensorimotor coordination adapts and develops by actions of sensory and motor experience. For tasks in which a-priori knowledge is limited, self-organising controllers are proposed. A novel technique is described, that allows a robot to autonomously determine its sensorimotor mapping. Artificial neural networks are used to learn the sensorimotor coordination, trained using self-organizing learning algorithms. The inverse kinematic mapping is learned by the use of a series of trial movements. The relation of the proposed algorithm to biological motion control is discussed.

A Neural Network Model with Self-organizing Pulse-Coupled Oscillator**The project supported by National Basic Research Project “Nonlinear Science”

We consider a two layered neural network in which the second layer is composed of simple pulse-coupled neuronal oscillators connected by excitatory and inhibitory lateral interaction weights. For storing information' a self-organizing learning algorithm with respect to the amplitude of the neuronal oscillators is introduced. The dynarnical behavior and the retrieval ability of information of this model are analyzed. Computer simulation shows that the learning can enhance the coherent characteristic of the neuronal activities and the dynamical behaviors of this model in pattern recognition are similar to the dynarnical process in biological neural system and perform information retrieval property to stored patterns.

Kalman Filter Implementation of Self-Organizing Feature Maps

The self-organizing learning algorithm of Kohonen and most of its extensions are controlled by two learning parameters, the learning coefficient and the width of the neighborhood function, which have to be chosen empirically because neither rules nor methods for their calculation exist. Consequently, often time-consuming parameter studies precede neighborhood-preserving feature maps of the learning data. To circumvent those lengthy numerical studies, this article describes the learning process by a state-space model in order to use the linear Kalman filter algorithm training the feature maps. Then the Kalman filter equations calculate the learning coefficient online during the training, while the width of the neighborhood function needs to be estimated by a second extended Kalman filter for the process of neighborhood preservation. The performance of the Kalman filter implementation is demonstrated on toy problems as well as on a crab classification problem. The results of crab classification are compared to those of generative topographic mapping, an alternative method to the self-organizing feature map.

Fuzzy classifications using fuzzy inference networks

In this paper, fuzzy inference models for pattern classifications have been developed and fuzzy inference networks based on these models are proposed. Most of the existing fuzzy rule-based systems have difficulties in deriving inference rules and membership functions directly from training data. Rules and membership functions are obtained from experts. Some approaches use backpropagation (BP) type learning algorithms to learn the parameters of membership functions from training data. However, BP algorithms take a long time to converge and they require an advanced setting of the number of inference rules. The work to determine the number of inference rules demands lots of experiences from the designer. In this paper, self-organizing learning algorithms are proposed for the fuzzy inference networks. In the proposed learning algorithms, the number of inference rules and the membership functions in the inference rules will be automatically determined during the training procedure. The learning speed is fast. The proposed fuzzy inference network (FIN) classifiers possess both the structure and the learning ability of neural networks, and the fuzzy classification ability of fuzzy algorithms. Simulation results on fuzzy classification of two-dimensional data are presented and compared with those of the fuzzy ARTMAP. The proposed fuzzy inference networks perform better than the fuzzy ARTMAP and need less training samples.

Self-organizing feature maps with self-adjusting learning parameters

This paper presents an extension of the self-organizing learning algorithm of feature maps in order to improve its convergence to neighborhood preserving maps. The Kohonen learning algorithm is controlled by two learning parameters, which have to be chosen empirically because there exists neither rules nor a method for their calculation. Consequently, often time consuming parameter studies have to precede before a neighborhood preserving feature map is obtained. To circumvent those lengthy numerical studies, here, a method is presented and incorporated into the learning algorithm which determines the learning parameters automatically. Therefore, system models of the learning and organizing process are developed in order to be followed and predicted by linear and extended Kalman filters. The learning parameters are optimal within the system models, so that the self-organizing process converges automatically to a neighborhood preserving feature map of the learning data.

On self-organizing algorithms and networks for class-separability features

We describe self-organizing learning algorithms and associated neural networks to extract features that are effective for preserving class separability. As a first step, an adaptive algorithm for the computation of Q(-1/2) (where Q is the correlation or covariance matrix of a random vector sequence) is described. Convergence of this algorithm with probability one is proven by using stochastic approximation theory, and a single-layer linear network architecture for this algorithm is described, which we call the Q(-1/2) network. Using this network, we describe feature extraction architectures for: 1) unimodal and multicluster Gaussian data in the multiclass case; 2) multivariate linear discriminant analysis (LDA) in the multiclass case; and 3) Bhattacharyya distance measure for the two-class case. The LDA and Bhattacharyya distance features are extracted by concatenating the Q (-1/2) network with a principal component analysis network, and the two-layer network is proven to converge with probability one. Every network discussed in the study considers a flow or sequence of inputs for training. Numerical studies on the performance of the networks for multiclass random data are presented.

Hierarchical fuzzy modelling for monitoring depth of anaesthesia

A hierarchical structure, based on fuzzy modelling, which can monitor depth of anaesthesia (DOA) from many clinical signs in the operating theatre, is described. The first level uses numerical clinical signs, such as systolic arterial pressure (SAP) and heart rate (HR), via a rule-base from anaesthetists' experience or a self-organizing learning algorithm to interpret primary depth of anaesthesia (PDOA). The second level is focused on non-numerical clinical signs, such as sweating (SW), lacrimation (LA) and pupil response (PR), which can be merged with the first level of PDOA to decide DOA with more confidence. Furthermore, linguistic rules and a simple fuzzy modelling concept have been used to model a patient during induction and maintenance stages. Not only can this model simulate new ideas and monitoring methods, but it can also be used clinically on-line to give an estimate of the adequacy of DOA. Successful results have given confidence to perform on-line clinical trials in the operating theatre. Two configurations of fuzzy modelling to determine the rule-base are described in this paper. Firstly, a self-organizing fuzzy modelling approach has the ability to generate rules on-line from input and output data. Secondly, if there is no direct measurement for the system output or it is very difficult to interpret, a self-organizing learning system which can learn rules from off-line input and output data can be utilized for modelling such systems.

An Information-Maximization Approach to Blind Separation and Blind Deconvolution

We derive a new self-organizing learning algorithm that maximizes the information transferred in a network of nonlinear units. The algorithm does not assume any knowledge of the input distributions, and is defined here for the zero-noise limit. Under these conditions, information maximization has extra properties not found in the linear case (Linsker 1989). The nonlinearities in the transfer function are able to pick up higher-order moments of the input distributions and perform something akin to true redundancy reduction between units in the output representation. This enables the network to separate statistically independent components in the inputs: a higher-order generalization of principal components analysis. We apply the network to the source separation (or cocktail party) problem, successfully separating unknown mixtures of up to 10 speakers. We also show that a variant on the network architecture is able to perform blind deconvolution (cancellation of unknown echoes and reverberation in a speech signal). Finally, we derive dependencies of information transfer on time delays. We suggest that information maximization provides a unifying framework for problems in "blind" signal processing.

A Gaussian potential function network with hierarchically self-organizing learning

This article presents a design principle of a neural network using Gaussian activation functions, referred to as a Gaussian Potential Function Network (GPFN), and explores the capability of a GPFN in learning a continuous input-output mapping from a given set of teaching patterns. The design principle is highlighted by a Hierarchically Self-Organizing Learning (HSOL) algorithm featuring the automatic recruitment of hidden units under the paradigm of hierarchical learning.A GPFN generates an arbitrary shape of a potential field over the domain of the input space, as an input-output mapping, by synthesizing a number of Gaussian potential functions provided by individual hidden units referred to as Gaussian Potential Function Units (GPFUs). The construction of a GPFN is carried out by the HSOL algorithm which incrementally recruits the minimum necessary number of GPFUs based on the control of the effective radii of individual GPFUs, and trains the locations (mean vectors) and shapes (variances) of individual Gaussian potential functions, as well as their summation weights, based on the Backpropagation algorithm.Simulations were conducted for the demonstration and evaluation of the GPFNs constructed based on the HSOL algorithm for several sets of teaching patterns.