On the Characteristics of Growing Cell Structures (GCS) Neural Network |
| |
Authors: | Wang Jung-Hua Sun Wei-Der |
| |
Affiliation: | (1) Department of Electrical Engineering, National Taiwan Ocean University, 2 Peining Rd, Keelung, Taiwan, e-mail |
| |
Abstract: | In this paper, a self-developing neural network model, namely the Growing Cell Structures (GCS) is characterized. In GCS each node (or cell) is associated with a local resource counter (t). We show that GCS has the conservation property by which the summation of all resource counters always equals
, thereby s is the increment added to (t) of the wining node after each input presentation and (0 < < 1.0) is the forgetting (i.e., decay) factor applied to (t) of non-wining nodes. The conservation property provides an insight into how GCS can maximize information entropy. The property is also employed to unveil the chain-reaction effect and race-condition which can greatly influence the performance of GCS. We show that GCS can perform better in terms of equi-probable criterion if the resource counters are decayed by a smaller . |
| |
Keywords: | self-developing neural network competitive learning race-condition topology equi-probable criterion chain-reaction effect |
本文献已被 SpringerLink 等数据库收录! |
|