Statistical Mechanics of Adaptive Weight Perturbation Learning

スポンサーリンク

概要

• 論文の詳細を見る

• Weight perturbation learning was proposed as a learning rule in which perturbation is added to the variable parameters of learning machines. The generalization performance of weight perturbation learning was analyzed by statistical mechanical methods and was found to have the same asymptotic generalization property as perceptron learning. In this paper we consider the difference between perceptron learning and AdaTron learning, both of which are well-known learning rules. By applying this difference to weight perturbation learning, we propose adaptive weight perturbation learning. The generalization performance of the proposed rule is analyzed by statistical mechanical methods, and it is shown that the proposed learning rule has an outstanding asymptotic property equivalent to that of AdaTron learning.

• （社）電子情報通信学会の論文
• 2011-10-01

著者

• Miyoshi Seiji

  The Faculty Of Engineering Science Kansai University

• Maeda Yutaka

  The Faculty Of Engineering Science Kansai University

• MIYOSHI Ryosuke

  the Graduate School of Science and Engineering, Kansai University

• Miyoshi Ryosuke

  The Graduate School Of Science And Engineering Kansai University

関連論文

• Statistical Mechanics of On-Line Learning Using Correlated Examples
• Statistical Mechanics of Adaptive Weight Perturbation Learning

スポンサーリンク