Convergent Superiority of Bipolar Nets from the Viewpoint of Input Set Separability

DOI Web Site 4 Citations 10 References
  • SHIRATSUCHI Hiroshi
    Faculty of Computer Science and Systems Engineering, Kyushu Institute of Technology
  • GOTANDA Hiromu
    Faculty of Engineering, Kinki University in Kyushu
  • INOUE Katsuhiro
    Faculty of Computer Science and Systems Engineering, Kyushu Institute of Technology
  • KUMAMARU Kousuke
    Faculty of Computer Science and Systems Engineering, Kyushu Institute of Technology

Bibliographic Information

Other Title
  • 入力集合の可分離性からみた両極型ネットの収束上の優位性
  • ニュウリョク シュウゴウ ノ カブンリセイ カラ ミタ リョウキョクガタ ネッ

Search this article

Description

In this paper, the separability condition that the separation hyperplanes pass through rectangles surrounding the input sets is formulated by normal vectors. Then distributions of the normal vectors satisfying the condition are depicted for the two-dimensional case. These distributions elucidate that the condition for the first hidden layer varies significantly even when the input patterns are simply translated, and that the conditions (in a wider sense) for the other layers are different depending on whether the unit is unipolar or bipolar, i.e. whether it activates from 0 to 1 or from-0.5 to 0.5. The initial distributions of the normal vectors with the weights initialized ordinarily by random numbers with zero mean are also depicted. Comparison of the initial distributions to the separability conditions leads to the conclusion that the bipolar nets are superior to the unipolar nets in convergence of the back propagation learning initialized in such an ordinary manner. The bipolar nets exhibit better convergence than the unipolar nets even if their input space divisions in every layer by the separation hyperplanes are geometrically the same at the outset of the learning.

Journal

Citations (4)*help

See more

References(10)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top