Studies on Effects of Initialization on Structure Formationand Generalization of Structural Learning with Forgetting
Hiroshi Shiratsuchi*, Hiromu Gotanda**, Katsuhiro Inoue***, and Kousuke Kumamaru***
*Faculty of Engineering, University of the Ryukyus, Nishihara, Okinawa 903-0213, Japan
**Kinki University School of Humanity-Oriented Science and Engineering
***Faculty of Computer Science and Systems Engineering, Kyushu Institute of Technology
In this paper, our proposed initialization for multilayer neural networks (NN) applies to the structural learning with forgetting. Initialization consists of two steps: weights of hidden units are initialized so that their hyperplanes pass through the center of gravity of an input pattern set, and weights of output units are initialized to zero. Several simulations were performed to study how the initialization effects the structure formation of the NN. From the simulation result, it was confirmed that the initialization gives better network structure and higher generalization ability.
This article is published under a Creative Commons Attribution-NoDerivatives 4.0 Internationa License.