ICANN 2009: Artificial Neural Networks – ICANN 2009 pp 381-390 | Cite as
Training Recurrent Neural Network Using Multistream Extended Kalman Filter on Multicore Processor and Cuda Enabled Graphic Processor Unit
Abstract
Recurrent neural networks are popular tools used for modeling time series. Common gradient-based algorithms are frequently used for training recurrent neural networks. On the other side approaches based on the Kalman filtration are considered to be the most appropriate general-purpose training algorithms with respect to the modeling accuracy. Their main drawbacks are high computational requirements and difficult implementation. In this work we first provide clear description of the training algorithm using simple pseudo-language. Problem with high computational requirements is addresses by performing calculation on Multicore Processor and CUDA-enabled graphic processor unit. We show that important execution time reduction can be achieved by performing computation on manycore graphic processor unit.
Keywords
Weight Connection Recurrent Neural Network Hide Unit Output Unit Previous Time StepPreview
Unable to display preview. Download preview PDF.
References
- 1.Werbos, P.: Backpropagation through time; what it does and how to do it. Proceedings of the IEEE 78, 1550–1560 (1990)CrossRefGoogle Scholar
- 2.Williams, R.J., Zipser, D.: A learning algorithm for continually running fully recurrent neural networks. Neural Computation 1, 270–280 (1989)CrossRefGoogle Scholar
- 3.Williams, R.J.: Some observations on the use of the extended Kalman filter as a recurrent network learning algorithm. Technical Report NU-CCS-92-1, Northeastern University, College of Computer Science, Boston, MA (1992)Google Scholar
- 4.Čerňanský, M., Beňušková, Ľ.: Simple recurrent network trained by RTRL and extended Kalman filter algorithms. Neural Network World 13(3), 223–234 (2003)Google Scholar
- 5.Trebatický, P.: Recurrent neural network training with the kalman filter-based techniques. Neural network world 15(5), 471–488 (2005)Google Scholar
- 6.Feldkamp, L., Prokhorov, D., Eagen, C., Yuan, F.: Enhanced multi-stream Kalman filter training for recurrent networks. In: Suykens, J., Vandewalle, J. (eds.) Nonlinear Modeling: Advanced Black-Box Techniques, pp. 29–53. Kluwer Academic Publishers, Dordrecht (1998)CrossRefGoogle Scholar
- 7.Prokhorov, D.V.: Toyota prius hev neurocontrol and diagnostics. Neural Networks 21, 458–465 (2008)CrossRefGoogle Scholar
- 8.NVIDIA: NVIDIA CUDA programming guide. Technical report (2008)Google Scholar
- 9.Trebatický, P.: Neural network training with extended kalman filter using graphics processing unit. In: Kůrková, V., Neruda, R., Koutník, J. (eds.) ICANN 2008, Part II. LNCS, vol. 5164, pp. 198–207. Springer, Heidelberg (2008)CrossRefGoogle Scholar
- 10.Prokhorov, D.V.: Kalman filter training of neural networks: Methodology and applications. In: Tutorial on IJCNN 2004, Budapest, Hungary (2004)Google Scholar
- 11.Elman, J.L.: Finding structure in time. Cognitive Science 14, 179–211 (1990)CrossRefGoogle Scholar
- 12.Elman, J.: Distributed representations, simple recurrent networks, and grammatical structure. Machine Learning 7, 195–225 (1991)Google Scholar