Parallel Batch Training of the Self-Organizing Map Using OpenCL
Lecture notes in computer sciencePublished 1 January 2010
Masahiro Takatsuka, Michael Bui
Citations13
SJR quartileQ2
SJR score0.35
SNIP0.55
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
This article presents the parallel implementation of SOMs, particularly the batch map variant using Graphics Processing Units (GPUs) through the use of Open Computing Language (OpenCL).
Abstract
The Self-Organizing Maps (SOMs) are popular artificial neural networks that are often used for data analyses through clustering and visualisation. SOM's mathematical model is inherently parallel. However, many implementations have not successfully exploited its parallelism because previous attempts often required cluster-like infrastructures. This article presents the parallel implementation of SOMs, particularly the batch map variant using Graphics Processing Units (GPUs) through the use of Open Computing Language (OpenCL).
Keywords
Computer Science
The OpenCL specification
634 Citations2009Aaftab Munshi
The specification is divided into a core specification that any OpenCL compliant implementation must support; a handheld/embedded profile which relaxes the OpenCL compliance requirements for handheld and embedded devices; and a set of optional extensions that are likely to move into the core specification in later revisions of the Opencl specification.
IEEE Transactions on Neural NetworksPicSOM-self-organizing image retrieval with MPEG-7 content descriptors
193 Citations2002Jorma Laaksonen, Markus Koskela +1 more
The results of the experiments show that the MPEG-7-defined content descriptors can be used as such in thePicSOM system even though Euclidean distance calculation, inherently used in the PicSom system, is not optimal for all of them.
Neural Network Implementation Using CUDA and OpenMP
135 Citations2008Hong-Hoon Jang, Anjin Park +1 more
This paper proposes more quick and efficient implementation of neural networks on both GPU and multi-core CPU and uses CUDA (compute unified device architecture) that can be easily programmed due to its simple C language-like style instead of GPU to solve the first problem.
Lecture notes in computer scienceSOFSEM 2002: Theory and Practice of Informatics
129 Citations2002Václav Hlaváč, Keith Jeffery
An overview on techniques for conformance testing of timed systems is given, some important models used to specify timed systems are described and the main techniques used for testing such systems are shown.
Neural NetworksSpherical self-organizing map using efficient indexed geodesic data structure
63 Citations2006Yingxin Wu, Masahiro Takatsuka
This work introduces a 2D rectangular grid data structure to store the icosahedron-based geodesic dome and develops an interface to project the GeoSOM onto the 2D plane using a cartographic approach, which gives users a global view of the spherical data map.
The European Symposium on Artificial Neural NetworksSelf-Organizing Maps Computing on Graphic Process Unit
25 Citations2005Zhongwen Luo, Hongzhi Liu +2 more
The result shows that GPU can make SOM computing much faster than standard CPU and it is reasonable to expect that graphic hardware will widely used in other ANN computing for getting high-performance.
Exploratory geospatial analysis using GeoVISTA Studio: from a desktop to the Web
15 Citations2002Masahiro Takatsuka, Mark Gahegan
This paper illustrates various features of GeoVISTA Studio, such as ease of program construction (visual programming), an open (non-proprietary) architecture, simple component-based integration and advanced deployment methods, which allow us to rapidly construct and share applications that combine geocomputation and geovisualization methods.
NeurocomputingA parallel general implementation of Kohonen's self-organizing map algorithm: performance and scalability
9 Citations2002Piotr Ożdżyński, Andy Lin +2 more
It is reported that a parallel implementation of the self-organizing map algorithm on a Beowulf commodity-class computing cluster scales very favorably with the number of available nodes and greatly speeds the computation of medium-to-large-scale cortical maps.
Lecture notes in computer scienceParallelizing self-organizing maps
5 Citations1997David Štrupl, Roman Neruda
Several ways of parallelizing the self-organizing network (Kohonen maps) are studied on the BSP-like parallel machine model and optimal number of processors and criteria for choosing the right task decomposition are presented.