✦ LIBER ✦

Orthogonal considerations in the design of neural networks for function approximation

✍ Scribed by B. Francois

Publisher: Elsevier Science
Year: 1996
Tongue: English
Weight: 604 KB
Volume: 41
Category: Article
ISSN: 0378-4754
DOI: 10.1016/0378-4754(95)00062-3

No coin nor oath required. For personal study only.

✦ Synopsis

Two problems occur in the design of feedforward neural networks: the choice of the optimal architecture and the initialization. Generally, input and output data of a system (or a function) are measured and recorded. Then, experimenters wish to design a neural network to map exactly these output values.

By formulating this as a continuous approximation problem, this paper shows that the use of ortbogonal functions is a partial optimization in the choice of hidden functions.

Parameter's initialization is obtained by using the knowledge of input and output data in the calculation of a discrete approximation. The hidden weights are found by constructing orthogonal directions on which the input values are represented. The pseudo-inverse is used to determine output weights such that the Euclidean distances between neural responses and output values are minimised.

📜 SIMILAR VOLUMES

Properties and performance of orthogonal

Properties and performance of orthogonal neural network in function approximation

✍ Chieh F. Sher; Ching-Shiow Tseng; Chen-San Chen 📂 Article 📅 2001 🏛 John Wiley and Sons 🌐 English ⚖ 216 KB

Backpropagation neural network has been applied successfully to solving uncertain problems in many fields. However, unsolved drawbacks still exist such as the problems of local minimum, slow convergence speed, and the determination of initial weights and the number of processing elements. In this pa

Performance comparison between the train

Performance comparison between the training method and the numerical method of the orthogonal neural network in function approximation

✍ Chen-San Chen; Ching-Shiow Tseng 📂 Article 📅 2004 🏛 John Wiley and Sons 🌐 English ⚖ 355 KB

The orthogonal neural network is a recently developed neural network based on the properties of orthogonal functions. It can avoid the drawbacks of traditional feedforward neural networks such as initial values of weights, number of processing elements, and slow convergence speed. Nevertheless, it n

The errors of approximation for feedforw

The errors of approximation for feedforward neural networks in the metric

✍ Feilong Cao; Rui Zhang 📂 Article 📅 2009 🏛 Elsevier Science 🌐 English ⚖ 564 KB

The estimate for approximation error of

The estimate for approximation error of spherical neural networks

✍ Feilong Cao; Huazhong Wang; Shaobo Lin 📂 Article 📅 2011 🏛 John Wiley and Sons 🌐 English ⚖ 135 KB 👁 1 views

A unified approach for neural network-li

A unified approach for neural network-like approximation of non-linear functionals

✍ Tianping Chen 📂 Article 📅 1998 🏛 Elsevier Science 🌐 English ⚖ 56 KB

In this paper, we give a universal approach to approximation of non-linear functionals and so called myopic input-output maps by neural network-like architectures. Strong theorems on equi-uniform approximation to functionals in abstract spaces are given. As applications, theorems on identification o

A neural network for the optimization of

A neural network for the optimization of communications network design

✍ G. Vichniac; M.(Gard.ner) Lepp; M. Steenstrup 📂 Article 📅 1988 🏛 Elsevier Science 🌐 English ⚖ 95 KB