Two-step Modified SOM for Parallel Calculation
نویسندگان
چکیده
This paper presents a simple modification of classic Kohonen network (SOM), which allows parallel processing of input data vectors or partitioning the problem in case of insufficient memory for all vectors from the training set. The algorithm pre-selects potential centroids of data clusters and uses them as weight vectors in the final SOM network. We have demonstrated the usage of this algorithm on images as well as on two well-known datasets representing handwritten digits.
منابع مشابه
Enhanced Parallel Som Based on Heterogeneous System Platform
In this paper, we propose an enhanced parallel Selforganizing Map (SOM) framework based on heterogeneous system platform, specifically Central Processing Unit (CPU) and Graphic Processing Unit (GPU) soldered together on a single chip. The framework is to improve speed of parallel SOM using GPU since processing parallel SOM on GPU burden by communication latency due to isolate device architectur...
متن کاملNear-Minimum-Time Motion Planning of Manipulators along Specified Path
The large amount of computation necessary for obtaining time optimal solution for moving a manipulator on specified path has made it impossible to introduce an on line time optimal control algorithm. Most of this computational burden is due to calculation of switching points. In this paper a learning algorithm is proposed for finding the switching points. The method, which can be used for both ...
متن کاملA study of parallel neural networks
Parallel Self-Organizing Map (Parallel-SOM) is proposed to modify Self-Organizing Map for parallel computing environments. In this model, the conventional repeated learning procedure is modified to learn just once. The once learning manner is more similar to human learning and memorizing activities. During training, every connection between neurons of input and output layers is considered as an...
متن کاملqu an t - ph / 9 80 80 25 v 2 2 0 A ug 1 99 8 A Study of Parallel Self - Organizing Map
A new Self-Organizing Map is proposed for information parallel processing purpose. In this model, Parallel-SOM, there are two separate layers of neurons connected together. The number of neurons in both layer and connections between them is equal to the number of total elements of input signals. The weight updating is managed through a sequence of operations among some unitary transformation an...
متن کاملqu an t - ph / 9 80 80 25 17 A ug 1 99 8 A Study of Parallel Self - Organizing Map
A new Self-Organizing Map is proposed for information parallel processing purpose. In this model, Parallel-SOM, there are two separate layers of neurons connected together. The number of neurons in both layer and connections between them is equal to the number of total elements of input signals. The weight updating is managed through a sequence of operations among some unitary transformation an...
متن کامل