Pattern recognition by an adaptive process of sample set construction

Sebestyen, George

doi:10.1109/tit.1962.1057766

Cited by 60 publications

(13 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here the four elements are determined by minimization of the evaluation function as in Eq. (14). Since the evaluation function is multimodal with respect to cc 1 , cc 2 , cc 3 , and cc 4 , a global search is necessary.…”

Section: Center Determination Using Genetic Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Design of a blind spatial and temporal RBF equalizer using a genetic algorithm

2005

View full text Add to dashboard Cite

SUMMARYThis paper describes a new solution to channel intersymbol interference and nonlinear distortion by means of an adaptive array antenna and radial basis function (RBF) equalizer. The RBF equalizer can reduce the influence of intersymbol interference and additive noise in the digital communication system. It is also well known to eliminate nonlinear distortion effectively. However, conventional RBF equalizers including our previous proposed equalizer, a spatial and temporal RBF equalizer using adaptive array antenna, need training data to learn channel states, which incurs a decline in the data rate. In this paper, without training data, we construct a blind spatial and temporal RBF equalizer using genetic algorithm. We also describe a method of synthesizing Bayesian decision variables of RBF equalizers in order to obtain better performance. The RBF equalizer based on the proposed method can achieve a remarkable improvement in BER characteristics and its performance is demonstrated by computer simulations.

show abstract

Section: Center Determination Using Genetic Algorithmmentioning

confidence: 99%

“…When r j (k) is received at the j-th equalizer, let P j 1 (k) and P j 2 (k) denote the probabilities that the reconstructed signal û(k − d) is -1 and 1, respectively. Then the following relation can be derived, considering that the transmitted signal u(k) is an independent and random binary signal {±1} [14]:…”

Section: Synthesis Of Rbf Equalizersmentioning

confidence: 99%

Design of a blind spatial and temporal RBF equalizer using a genetic algorithm

2005

View full text Add to dashboard Cite

show abstract

“…Merge virtual data with original data : For estimation of the probability density function(PDF) of the training data set, we use a mixture gaussian kernels since we have no prior statistical information [6] [7] [8]. The familiarity of the j th kernel function to input x is thus defined as…”

Section: Network Combining Based On Cluster Distancementioning

confidence: 99%

Observational Learning with Modular Networks

Shin

Lee

Cho

2000

Intelligent Data Engineering and Automated Learning — IDEAL 2000. Data Mining, Financial Engineering, and Intelligent Agents

View full text Add to dashboard Cite

Abstract. Observational learning algorithm is an ensemble algorithm where each network is initially trained with a bootstrapped data set and virtual data are generated from the ensemble for training. Here we propose a modular OLA approach where the original training set is partitioned into clusters and then each network is instead trained with one of the clusters. Networks are combined with different weighting factors now that are inversely proportional to the distance from the input vector to the cluster centers. Comparison with bagging and boosting shows that the proposed approach reduces generalization error with a smaller number of networks employed.

show abstract

“…Sebestyen (Sebestyen, 1962) utilized a threshold based adaptive approach to determine the number of clusters. MacQueen's K-means algorithm (MacQueen, 1967) solves this issue by utilizing two external parameters to de ne the coarseness and re nement of the clustering.…”

Section: Scalementioning

confidence: 99%

An On-Line Agglomerative Clustering Method for Nonstationary Data

1999

View full text Add to dashboard Cite

An on-line agglomerative clustering algorithm for non-stationary data is described. Three issues are addressed. The rst regards the temporal aspects of the data. The clustering of stationary data by the proposed algorithm is comparable to the other popular algorithms tested (batch and on-line). The second issue addressed is the number of clusters required to represent the data. The algorithm provides an e cient framework to determine the natural number of clusters given the scale of the problem. Finally, the proposed algorithm implicitly minimizes the local distortion { a measure which t a k es into account clusters with relatively small mass.In contrast, most existing on-line clustering methods assume stationarity o f t h e d a t a . W h e n used to cluster non-stationary data these methods fail to generate a good representation. Moreover, most current algorithms are computationally intensive when determining the correct number of clusters. These algorithms tend to neglect clusters of small mass due to their minimization of the global distortion (Energy).

show abstract

Pattern recognition by an adaptive process of sample set construction

Cited by 60 publications

References 0 publications

Design of a blind spatial and temporal RBF equalizer using a genetic algorithm

Design of a blind spatial and temporal RBF equalizer using a genetic algorithm

Observational Learning with Modular Networks

An On-Line Agglomerative Clustering Method for Nonstationary Data

Contact Info

Product

Resources

About