Theory of Recurrent Neural Network with Common Synaptic Inputs
-
- Kawamura Masaki
- Faculty of Science, Yamaguchi University
-
- Yamana Michiko
- System Engineering Research Laboratory, Central Research Institute of Electric Power Industry
-
- Okada Masato
- Department of Complexity Science and Engineering, Graduate School of Frontier Sciences, The University of Tokyo Laboratory for Mathematical Neuroscience, RIKEN Brain Science Institute “Intelligent Cooperation and Control,” PRESTO, JST
Search this article
Abstract
We discuss the effects of common synaptic inputs in a recurrent neural network. Because of the effects of these common synaptic inputs, the correlation between neural inputs cannot be ignored, and thus the network exhibits sample dependence. Networks of this type do not have well-defined thermodynamic limits, and self-averaging breaks down. We therefore need to develop a suitable theory without relying on these common properties. While the effects of the common synaptic inputs have been analyzed in layered neural networks, it was apparently difficult to analyze these effects in recurrent neural networks due to feedback connections. We investigated a sequential associative memory model as an example of recurrent networks and succeeded in deriving a macroscopic dynamical description as a recurrence relation form of a probability density function.
Journal
-
- Journal of the Physical Society of Japan
-
Journal of the Physical Society of Japan 74 (11), 2961-2965, 2005
THE PHYSICAL SOCIETY OF JAPAN
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1390001204186514688
-
- NII Article ID
- 110002325531
- 130004539021
- 210000105734
-
- NII Book ID
- AA00704814
-
- BIBCODE
- 2005JPSJ...74.2961K
-
- ISSN
- 13474073
- 00319015
-
- NDL BIB ID
- 7699559
-
- Text Lang
- en
-
- Data Source
-
- JaLC
- NDL
- Crossref
- CiNii Articles
- KAKEN
-
- Abstract License Flag
- Disallowed