Abstract
In this paper we present an approach to the training of feed forward neural networks on massively parallel SIMD-architectures. In order to cover a wide field of applications we focus our attention on the flexibility of the load balancing routines. Our approach is characterized by three important properties: 1. All four types of parallelism inherent in the training phase are used. 2. In a preprocessing step neural networks are transformed into equivalent topologies, more suited for parallel computation. 3. Each learning task can be parallelized in a number of different ways, the best of which is chosen according to estimations of the computing efficiency.
Following these concepts we developed PINK2, a massively parallel simulator kernel for the MasPar MP1216. In contrast to most known approaches, efficient only for special topologies, it achieves good computing performance on a broad range of differing benchmark problems.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
A. Singer: Parallel Computing 14, 305 (1990)
X. Zhang, M. Mckenna, J.P. Mesirov, D.L. Waltz: NIPS 2, 801 (1990)
Wei-Ming Lin, V.K. Prasanna, K.W. Przytula: IEEE Transactions on Computers, Vol. 40, No. 12, 1390 (Dec 1991)
N. Mache, Master thesis, University of Stuttgart, 1992
I. Pitas (ed), A. Petrowski, H. Paugam-Moisy: Parallel Algorithms pp 259–328 Wiley 1993
S.E. Fahlmann: CMU-CS-88-162 (1988)
M. Riedmiller, H. Braun: Proc. ICNN ’93, 379 (1993)
H. Braun, J. Weisbrod: Proc. ICANNGA ’93, 25 (1993)
J. Schaefer, H. Braun: Proc. ICANNGA ’95
H. Braun, P. Zagorski: Proc. 3rd PPSN, 444 (1994)
W. Butscher: Supercomputer ’91, 187 (1991)
A. Zell: Simulation Neuronaler Netze, Addison Wesley 1994
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1995 Springer-Verlag/Wien
About this paper
Cite this paper
Koll, D., Riedmiller, M., Braun, H. (1995). Massively Parallel Training of Multi Layer Perceptrons With Irregular Topologies. In: Artificial Neural Nets and Genetic Algorithms. Springer, Vienna. https://doi.org/10.1007/978-3-7091-7535-4_77
Download citation
DOI: https://doi.org/10.1007/978-3-7091-7535-4_77
Publisher Name: Springer, Vienna
Print ISBN: 978-3-211-82692-8
Online ISBN: 978-3-7091-7535-4
eBook Packages: Springer Book Archive