Building Recurrent Neural Networks to Implement Multiple Attractor Dynamics Using the Gradient Descent Method. (27th October 2008)

• Record Type: Journal Article
• Authors: Namikawa, Jun; Tani, Jun
• Other Names: Imada Akira Academic Editor.
• Is Part Of: Advances in artificial neural systems. (2009)
• Publication Date: 2008-10-27
• Subjects: Neural networks (Computer science) -- Periodicals; Neural networks (Computer science); Periodicals; Electronic journals; 006.32
• DOI: 10.1155/2009/846040
• Languages: English
• ISSNs: 1687-7594

Abstract

Abstract : 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.