![](/images/graphics-bg.png)
Building Recurrent Neural Networks to Implement Multiple Attractor Dynamics Using the Gradient Descent Method
Joint Authors
Source
Advances in Artificial Neural Systems
Issue
Vol. 2009, Issue 2009 (31 Dec. 2009), pp.1-11, 11 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2008-10-27
Country of Publication
Egypt
No. of Pages
11
Main Subjects
Information Technology and Computer Science
Abstract EN
The present paper proposes a recurrent neural network model and learning algorithm that can acquire the ability to generate desired multiple sequences.
The network model is a dynamical system in which the transition function is a contraction mapping, and the learning algorithm is based on the gradient descent method.
We show a numerical simulation in which a recurrent neural network obtains a multiple periodic attractor consisting of five Lissajous curves, or a Van der Pol oscillator with twelve different parameters.
The present analysis clarifies that the model contains many stable regions as attractors, and multiple time series can be embedded into these regions by using the present learning method.
American Psychological Association (APA)
Namikawa, Jun& Tani, Jun. 2008. Building Recurrent Neural Networks to Implement Multiple Attractor Dynamics Using the Gradient Descent Method. Advances in Artificial Neural Systems،Vol. 2009, no. 2009, pp.1-11.
https://search.emarefa.net/detail/BIM-502854
Modern Language Association (MLA)
Namikawa, Jun& Tani, Jun. Building Recurrent Neural Networks to Implement Multiple Attractor Dynamics Using the Gradient Descent Method. Advances in Artificial Neural Systems No. 2009 (2009), pp.1-11.
https://search.emarefa.net/detail/BIM-502854
American Medical Association (AMA)
Namikawa, Jun& Tani, Jun. Building Recurrent Neural Networks to Implement Multiple Attractor Dynamics Using the Gradient Descent Method. Advances in Artificial Neural Systems. 2008. Vol. 2009, no. 2009, pp.1-11.
https://search.emarefa.net/detail/BIM-502854
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-502854