A Constructive Algorithm to Determine the Feasibility and Weights of Two-Layer Perceptrons for Celled Decision Regions

نویسنده

  • CHE-CHERN LIN
چکیده

Necessary and sufficient conditions for implementing particular decision regions by multi-layer perceptrons have been presented in recent studies. In this paper, from a viewpoint of engineering, a constructive algorithm is proposed to implement celled decision regions using two-layer perceptrons without any training procedure. The algorithm examines the feasibility of a celled decision region and then determines the weights of the second layer for a particular two-layer perceptron to implement the decision region if it is realizable. The algorithm is fast based on two aspects. First, it tests the feasibility and determines the weights without any training procedure. Second, the classifications of input patterns are based on integer manipulations since the weights determined by the algorithm are all integers. The proposed algorithm consists of only three simple steps and is implemented easily by computer programming languages. Key-Words: Classification; Multi-layer perceptron; Celled decision region; Partitioning capability.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

On Implementation of Nested Rectangular Decision Regions by Multi-layer Perceptrons I: Algorithm

There are three papers in a series of discussions related to the partitioning capabilities on nested rectangular decision regions using multi-layer perceptrons. We propose a constructive algorithm, called the up-down algorithm, to realize the nested rectangular decision regions. The algorithm determines the weights easily and can be generalized to solve other decision regions with the propertie...

متن کامل

Implementation Feasibility of Convex Recursive Deletion Regions Using Multi-Layer Perceptrons

A constructive algorithm to implement convex recursive deletion regions via two-layer perceptrons has been presented in a recent study. In the algorithm, the absolute values of the weights become larger and larger when the number of nested layers of a convex recursive deletion region increases. In addition, the absolute values of the weights are determined according to the complexity of the str...

متن کامل

Modeling of measurement error in refractive index determination of fuel cell using neural network and genetic algorithm

Abstract: In this paper, a method for determination of refractive index in membrane of fuel cell on basis of three-longitudinal-mode laser heterodyne interferometer is presented. The optical path difference between the target and reference paths is fixed and phase shift is then calculated in terms of refractive index shift. The measurement accuracy of this system is limited by nonlinearity erro...

متن کامل

Neural Net and Traditional Classifiers

Previous work on nets with continuous-valued inputs led to generative procedures to construct convex decision regions with two-layer perceptrons (one hidden layer) and arbitrary decision regions with three-layer perceptrons (two hidden layers). Here we demonstrate that two-layer perceptron classifiers trained with back propagation can form both convex and disjoint decision regions. Such classif...

متن کامل

An algorithm for determining common weights by concept of membership function

Data envelopment analysis (DEA) is a method to evaluate the relative efficiency of decision making units (DMUs). In this method, the issue has always been to determine a set of weights for each DMU which often caused many problems. Since the DEA models also have the multi-objective linear programming (MOLP) problems nature, a rational relationship can be established between MOLP and DEA problem...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2006