Modularity in neural computing

Caelli, Terry; Guan, Ling; Wang, Wen

doi:10.1109/5.784227

Cited by 66 publications

(25 citation statements)

References 116 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A modular neural network (MNN) is a neural network that consists of several single neural network models, each one solves one subproblem of the original problem, and the solutions of models are integrated to form the solution of the original problem (Caelli et al 1999). Figure 1(b) shows the general modular neural network structure using three neural network models.…”

Section: Modular Neural Networkmentioning

confidence: 99%

“…The modular neural network can increase the computational efficiency of solving large size/ high complex problems, by decomposing it into several subproblems with small sizes and each is handled by an efficient neural network model. But, how to divide a problem into subproblems and how to assign a neural network model to learn the subproblems are often difficult for the combination of these submodel outputs to form the global optimal solution (Jacobs et al 1991;Anand et al 1995;Copeland 1995;Kecman 1996;Kehagias and Petridis 1997;Kadirkamanathan and Fabri 1998;Yamaguchi and Itakura 1999;Hashem 1997;Sierra and Cruz 1998;Caelli et al 1999). To overcome these drawbacks, cooperative modular neural network approaches were proposed (Auda and Kamel 1997a).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cooperative recurrent modular neural networks for constrained optimization: a survey of models and applications

Kamel

Xia

2008

Cogn Neurodyn

View full text Add to dashboard Cite

Constrained optimization problems arise in a wide variety of scientific and engineering applications. Since several single recurrent neural networks when applied to solve constrained optimization problems for real-time engineering applications have shown some limitations, cooperative recurrent neural network approaches have been developed to overcome drawbacks of these single recurrent neural networks. This paper surveys in details work on cooperative recurrent neural networks for solving constrained optimization problems and their engineering applications, and points out their standing models from viewpoint of both convergence to the optimal solution and model complexity. We provide examples and comparisons to shown advantages of these models in the given applications.

show abstract

Section: Modular Neural Networkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Cooperative recurrent modular neural networks for constrained optimization: a survey of models and applications

Kamel

Xia

2008

Cogn Neurodyn

View full text Add to dashboard Cite

show abstract

“…The first group of these algorithms relates to techniques where the structures of the connected submodels are predetermined before learning (Caelli et al, 1999). The paper highlights that the biological neural networks are relatively rare connected inspiring the necessity of distributed modelling solutions and that the assignment solutions of various applications can be quicker having a computer model consisting of net of submodels structure built up usually on the base of field specific know-how.…”

Section: Approaches For Ann Based Decision Support Structuresmentioning

confidence: 99%

Automatic Generation a Net of Models for High and Low Levels of Production Control

Viharos

2005

IFAC Proceedings Volumes

View full text Add to dashboard Cite

The paper presents two applications of the novel, artificial neural network (ANN) and feature selection based combined, dynamic technique to automatically dissolve a large, complex system into a net of connected submodels. The first application is a solution for the lower level of customised mass-production systems, for increasing their productivity. The second one is a concept for the identification and reorganisation of manufacturing agents, based on simulation experience. The main idea behind, is to give the learning capability to agents already at their definition phase, and to maximise their foresight power.

show abstract

“…For example, in Ref. 6, Caelli, Guan, and Wen studied neural network in terms of collections of subnetwork module. The result indicated that modular neural network can reduce the computational complexity.…”

Section: Introductionmentioning

confidence: 99%

Modelling a neural network using an algebraic method

Sugunnasil¹,

Somhom²,

Jumpamule³

et al. 2014

ScienceAsia

View full text Add to dashboard Cite

In this paper, a framework based on algebraic structures to formalize various types of neural networks is presented. The working strategy is to break down neural networks into building blocks, relationships between each building block, and their operations. Building blocks are collections of primary components or neurons. In turn, neurons are collections of properties functioning as single entities, transforming an input into an output. We perceive a neuron as a function. Thus the flow of information in a neural network is a composition between functions. Moreover, we also define an abstract data structure called a layer which is a collection of entities which exist in the same time step. This layer concept allows the parallel computation of our model. There are two types of operation in our model; recalling operators and training operators. The recalling operators are operators that challenge the neural network with data. The training operators are operators that change parameters of neurons to fit with the data. This point of view means that all neural networks can be constructed or modelled using the same structures with different parameters.

show abstract

Modularity in neural computing

Abstract: This paper considers neural computing models for information processing in terms of

Cited by 66 publications

References 116 publications

Cooperative recurrent modular neural networks for constrained optimization: a survey of models and applications

Cooperative recurrent modular neural networks for constrained optimization: a survey of models and applications

Automatic Generation a Net of Models for High and Low Levels of Production Control

Modelling a neural network using an algebraic method

Contact Info

Product

Resources

About