Name
Title | ||
|---|---|---|
Reducing Feed-Forward Neural Network Processing Time Utilizing Matrix Multiplication Algorithms on Heterogeneous Distributed Systems |
Abstract | ||
|---|---|---|
This paper presents a work in progress that aims to reduce the overall training and processing time of feed-forward multi-layer neural networks. If the network is large processing is expensive in terms of both; time and space. In this paper, we suggest a cost-effective and presumably a faster processing technique by utilizing a heterogeneous distributed system composed of a set of commodity computers connected by a local area network. Neural network computations can be viewed as a set of matrix multiplication processes. These can be adapted to utilize the existing matrix multiplication algorithms tailored for such systems. With Java technology as an implementation means, we discuss the different factors that should be considered in order to achieve this goal highlighting some issues that might affect such a proposed implementation. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1109/CICSYN.2009.67 | CICSyN |
Keywords | Field | DocType |
proposed implementation,processing time,time utilizing matrix multiplication,large processing,feed-forward multi-layer neural network,faster processing technique,existing matrix multiplication,java technology,matrix multiplication process,feed-forward neural network processing,neural network computation,local area network,benchmark testing,feedforward neural networks,hardware,neural network,cost effectiveness,distributed computing,local area networks,neural networks,feed forward,artificial neural networks,java,computer networks,computational modeling,feed forward neural network,computational intelligence,matrix multiplication,work in progress,parallel processing | Feedforward neural network,Computational intelligence,Work in process,Computer science,Algorithm,Local area network,Artificial neural network,Matrix multiplication,Benchmark (computing),Computation,Distributed computing | Conference |
ISBN | Citations | PageRank |
978-0-7695-3743-6 | 0 | 0.34 |
References | Authors | |
13 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ali Kattan | 1 | 23 | 2.46 |
| Rosni Abdullah | 2 | 156 | 24.82 |
| Rosalina Abdul Salam | 3 | 70 | 10.83 |