Chinese character learning by synchronization in Wilson-Cowan oscillatory neural networks

  • Authors:

  • Jiaxin Cui;Yan Liu;Jiawei Chen;Liujun Chen;Fukang Fang

  • Affiliations:

  • Department of Systems Science, School of Management, Beijing Normal University, Beijing, P.R. China;Department of Systems Science, School of Management, Beijing Normal University, Beijing, P.R. China;Department of Systems Science, School of Management, Beijing Normal University, Beijing, P.R. China;Department of Systems Science, School of Management, Beijing Normal University, Beijing, P.R. China;Institute of Nonequilibrium System, Beijing Normal University, Beijing, P.R. China

  • Venue:
  • ICNC'09 Proceedings of the 5th international conference on Natural computation
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

The cognition and learning of Chinese characters is a good example of human visual perception and learning. The synchronization behavior between neuronal groups in cortical areas is one of the core mechanisms in visual image perception and recognition. We built a neural network with locally coupled Wilson-Cowan oscillators to learn Chinese characters. Neurons in each group representing the same feature of a character will be synchronized so that the common features of Chinese characters will be extracted by the network, such as a common character in two-character words or a common radical component of characters. Further more, each feature is represented by a stable fixed point attractor in the dynamic neural network.