 |
[email protected] |
 |
3275638434 |
 |
 |
| Paper Publishing WeChat |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Neural Network Inversion for Multilayer Quaternion Neural Networks
Takehiko Ogawa
Full-Text PDF
XML 1870 Views
DOI:10.17265/1934-7332/2016.02.002
Department of Electronics and Computer Systems, Faculty of Engineering, Takushoku University, Tokyo 193-0985, Japan
Recently, solutions to inverse problems have been required in various engineering fields. The neural network inversion method has been studied as one of the neural network-based solutions. On the other hand, the extension of the neural network to a higher-dimensional domain, e.g., complex-value or quaternion, has been proposed, and a number of higher-dimensional neural network models have been proposed. Using the quaternion, we have the advantage of expressing 3D (three-dimensional) object attitudes easily. In the quaternion domain, we can define inverse problems where the cause and the result are expressed by the quaternion. In this paper, we extend the neural network inversion method to the quaternion domain. Further, we provide the results of the computer experiments to demonstrate the process and effectiveness of our method.
Inverse problems, neural network inversion, quaternion, inverse mapping, inverse kinematics.
[2] Neto, F. D. M., and da Silva Neto, A. J. 2011. An Introduction to Inverse Problems with Applications. Springer.
[3] Linden, A., and Kindermann, J. 1989. “Inversion of Multilayer Nets.” In Proceedings of the International Joint Conference on Neural Networks, 425-30. Washington, D.C.
[4] Valova, I., Kameyama, K., and Kosugi, Y. 1995. “Image Decomposition by Answer-in-Weights Neural Network.” IEICE Transactions on Information and Systems E78-D (9): 1221-4.
[5] Lu, B., and Ito, K. 1995. “Regularization of Inverse Kinematics for Redundant Manipulators Using Neural Network Inversions.” In Proceedings of IEEE International Conference on Neural Networks, 2726-31. Perth, WA.
[6] Murray, W. R., Heg, C. T., and Pohlhammer, C. M. 1993. “Iterative Inversion of a Neural Network for Estimating the Location of a Planar Object.” Proceedings of the World Congress on Neural Networks 3: 188-93.
[7] Hirose, A. 2006. Complex-Valued Neural Networks. Springer.
[8] Nitta, T. 2009. Complex-Valued Neural Networks: Utilizing High-Dimensional Parameters. IGI-Global.
[9] Nitta, T. 1996. “An Extension of the Back-Propagation Algorithm to Quaternions.” Proceedings of the International Conference on Neural Information Processing 1: 247-50. Hong Kong.
[10] Matsui, N., Isokawa, T., Kusamichi, H., Peper, F., and Nishimura, H. 2004. “Quaternion Neural Network with Geometrical Operators.” Journal of Intelligent and Fuzzy Systems 15 (3-4): 149-64.
[11] Kuroe, Y. 2011. “Models of Clifford Recurrent Neural Networks and Their Dynamics.” Proceedings of the 2011 International Joint Conference on Neural Networks, 1035-41. San Jose.
[12] Ogawa, T. 2009. “Complex-Valued Neural Network and Inverse Problems.” In Complex-Valued Neural Networks: Utilizing High-Dimensional Parameters, edited by T. Nitta. IGI-Global, Chapter 2, 27-55.
[13] Fukami, S., Ogawa, T., and Kanada, H. 2008. “Regularization for Complex-Valued Network Inversion.” Proceedings of the SICE Annual Conference, 1237-42. Tokyo.
[14] Conway, J. H., and Smith, D. A. 2003. On Quaternions and Octonions—Their Geometry, Arithmetic, and Symmetry. A. K. Peters, Ltd.
[15] Garling, D. J. H. 2011. Clifford Algebras: An Introduction. Cambridge University Press.
[16] Kristiansen, R., and Nicklasson, P. J. 2005. “Satellite Attitude Control by Quaternion-Based Backstepping.” Proceedings of the 2005 American Control Conference 2: 907-12.
[17] Vince, J. 2010. Geometric Algebra for Computer Graphics. Springer.