A New Approach to the Structural Learning of Neural Networks(Neural Networks and Bioengineering)

スポンサーリンク

概要

• 論文の詳細を見る

• Structural learning algorithms are obtained by adding a penalty criterion (usually comes from the network structure) to the conventional criterion of the sum of squared errors and applying the backpropagation (BP) algorithm. This problem can be viewed as a constrained minimization problem. In this paper, we apply the Lagrangian differential gradient method to the structural learning based on the backpropagation-like algorithm. Computational experiments for both artificial and real data show that the improvement of generalization performance and the network optimization are obtained applying the proposed method.

• 社団法人電子情報通信学会の論文
• 2004-06-01

著者

• Debnath R

  Department Of Information And Communication Engineering The University Of Electro-communications

• TAKAHASHI Haruhisa

  Department of Information and Communication Engineering, The University of Electro-Communications

• Takahashi Haruhisa

  Department Of Information And Communication Engineering The University Of Electro-communications

• Takahashi Haruhisa

  Department Of Communications And Systems Engineering The University Of Electro-communications

• Debnath Rameswar

  Department Of Information And Communication Engineering The University Of Electro-communications

関連論文

• Implementation Issues of Second-Order Cone Programming Approaches for Support Vector Machine Learning Problems
• Kernel Selection for the Support Vector Machine(Biocybernetics, Neurocomputing)
• A New Approach to the Structural Learning of Neural Networks(Neural Networks and Bioengineering)
• Learning Curves in Learning with Noise : An Empirical Study
• Estimating Learning Curves of Concept Learning

スポンサーリンク