• Corpus ID: 14336869

Kohonen Feature Maps and Growing Cell Structures - a Performance Comparison

@inproceedings{Fritzke1992KohonenFM,
  title={Kohonen Feature Maps and Growing Cell Structures - a Performance Comparison},
  author={Bernd Fritzke},
  booktitle={NIPS},
  year={1992}
}
A performance comparison of two self-organizing networks, the Kohonen Feature Map and the recently proposed Growing Cell Structures is made. For this purpose several performance criteria for self-organizing networks are proposed and motivated. The models are tested with three example problems of increasing difficulty. The Kohonen Feature Map demonstrates slightly superior results only for the simplest problem. For the other more difficult and also more realistic problems the Growing Cell… 

Figures from this paper

Growing cell structures--A self-organizing network for unsupervised and supervised learning

Externally Growing Cell Structures for Pattern Classification

TLDR
Simulation results on some neural network benchmarks indicate that EGCS performs better than the original GCS, measured by classification rate, the required number of cells and epochs.

Growing mechanisms and cluster identification with TurSOM

TLDR
This work presents an application driven approach, based on the theoretical and empirical work of previous TurSOM experiments, to eliminate the need for post processing methods of cluster identification using SOM algorithms.

Growing Grid — a self-organizing network with constant neighborhood range and adaptation strength

TLDR
A novel self-organizing network which is generated by a growth process where both the neighborhood range used to co-adapt units in the vicinity of the winning unit and the adaptation strength are constant during the growth phase.

Cooling Growing Grid: An IncrementalSelf-organizing Neural Network ForData Exploration

TLDR
A novel incremental self-organizing ANN Cooling Growing Grid (CGG) is proposed, which combines the advantages of static and incremental approaches and overcomes their main drawbacks.

Identification of Patterns via Region-Growing Parallel SOM Neural Network

TLDR
An algorithm has been proposed to take advantage of the unique characteristics of the ParaSOM, where each neuron actually covers a region of the input space, and opens up possibilities for different approaches to clustering and classification.

Hierarchical growing cell structures: TreeGCS

  • Victoria J. HodgeJ. Austin
  • Computer Science
    KES'2000. Fourth International Conference on Knowledge-Based Intelligent Engineering Systems and Allied Technologies. Proceedings (Cat. No.00TH8516)
  • 2000
We propose a hierarchical, unsupervised clustering algorithm (TreeGCS) based upon the Growing Cell Structure (GCS) neural network of Fritzke. Our algorithm improves an inconsistency in the GCS

Comparison of dynamic feature map models for environmental monitoring

  • T. TrautmannT. Denoeux
  • Computer Science
    Proceedings of ICNN'95 - International Conference on Neural Networks
  • 1995
TLDR
Three topology feature map models with architecture adaptation are applied to water quality data and compared to the basic SOM model with fixed architecture using five objective performance criteria and the BungySOM model is shown to outperform the other constructive algorithms and to carry out faster adaptation to new inputs.

Dynamic Cell Structure Learns Perfectly Topology Preserving Map

TLDR
Simulations on a selection of CMU-Benchmarks indicate that the DCS idea applied to the growing cell structure algorithm leads to an efficient and elegant algorithm that can beat conventional models on similar tasks.
...

References

SHOWING 1-10 OF 13 REFERENCES

Unsupervised clustering with growing cell structures

  • B. Fritzke
  • Computer Science
    IJCNN-91-Seattle International Joint Conference on Neural Networks
  • 1991
TLDR
A neural network model is presented which is able to detect clusters of similar patterns and the accuracy of the cluster description increases linearly with the number of evaluated sample vectors.

Fast Learning in Networks of Locally-Tuned Processing Units

We propose a network architecture which uses a single internal layer of locally-tuned processing units to learn both classification tasks and real-valued function approximations (Moody and Darken

FLEXMAP-a neural network for the traveling salesman problem with linear time and space complexity

  • Bernd FritzkeP. Wilke
  • Computer Science
    [Proceedings] 1991 IEEE International Joint Conference on Neural Networks
  • 1991
TLDR
The authors present a self-organizing 'neural' network for the traveling salesman problem that is partly based on the model of Kohonen as no ring structure with a fixed number of elements is used.

Quantifying the neighborhood preservation of self-organizing feature maps

TLDR
It is shown that a topographic product P, first introduced in nonlinear dynamics, is an appropriate measure of the preservation or violation of neighborhood relations and it is found that a 3D output space seems to be optimally suited to the data.

Walker1991], \A study of the application of Kohonen-type neural networks to the travelling Salesman Problem

  • Biological Cybernetics

Self-Organizing Semantic Maps," Biological Cybernetics 61,241-254

  • T. Kohonen
  • 1989

Self-Organized Formation of Correct Feature Maps

  • 1982

A fully neural approach to image compression,

  • Artificial Neural Networks,
  • 1991