Graphics hardware implementation of the parameter-less self-organising map

Authors:
Alexander Campbell;Erik Berglund;Alexander Streit
Affiliations:
Faculty of IT, Queensland University of Technology, Brisbane, Australia;Information Technology and Electrical Engineering, University of Queensland, St. Lucia, Australia;Faculty of IT, Queensland University of Technology, Brisbane, Australia
Venue:
IDEAL'05 Proceedings of the 6th international conference on Intelligent Data Engineering and Automated Learning
Year:
2005

Citing 2
Cited 5

Neural computation and self-organizing maps: an introduction

Neural computation and self-organizing maps: an introduction
The parameterless self-organizing map algorithm

IEEE Transactions on Neural Networks

Browsing a Large Collection of Community Photos Based on Similarity on GPU

ISVC '08 Proceedings of the 4th International Symposium on Advances in Visual Computing, Part II
Fuzzy ART neural network parallel computing on the GPU

IWANN'07 Proceedings of the 9th international work conference on Artificial neural networks
Fuzzy ARTMAP based neural networks on the GPU for high-performance pattern recognition

IWINAC'11 Proceedings of the 4th international conference on Interplay between natural and artificial computation: new challenges on bioinspired applications - Volume Part II
Utilization of a rational-based representation to improve the image quality of a hardware-based K-SOM quantizer

Journal of Real-Time Image Processing
Similarity-based image organization and browsing using multi-resolution self-organizing map

Image and Vision Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a highly parallel implementation of a new type of Self-Organising Map (SOM) using graphics hardware. The Parameter-Less SOM smoothly adapts to new data while preserving the mapping formed by previous data. It is therefore in principle highly suited for interactive use, however for large data sets the computational requirements are prohibitive. This paper will present an implementation on commodity graphics hardware which uses two forms of parallelism to significantly reduce this barrier. The performance is analysed experimentally and algorithmically. An advantage to using graphics hardware is that visualisation is essentially “free”, thus increasing its suitability for interactive exploration of large data sets.