On-Line Learning Theory of Soft Committee Machines with Correlated Hidden Units-Steepest Gradient Descent and Natural Gradient Descent-

DOI Web Site Web Site BIBCODE 被引用文献6件 参考文献9件
  • Inoue Masato
    Laboratory for Mathematical Neuroscience, RIKEN Brain Science Institute Department of Otolaryngology, Head and Neck Surgery, Graduate School of Medicine, Kyoto University
  • Park Hyeyoung
    Laboratory for Mathematical Neuroscience, RIKEN Brain Science Institute
  • Okada Masato
    Laboratory for Mathematical Neuroscience, RIKEN Brain Science Institute “Intelligent Cooperation and Control”, PRESTO, JST

この論文をさがす

抄録

The permutation symmetry of the hidden units in multilayer perceptrons causes the saddle structure and plateaus of the learning dynamics in gradient learning methods. The correlation of the weight vectors of hidden units in a teacher network is thought to affect this saddle structure, resulting in a prolonged learning time, but this mechanism is still unclear. In this paper, we discuss it with regard to soft committee machines and on-line learning using statistical mechanics. Conventional gradient descent needs more time to break the symmetry as the correlation of the teacher weight vectors rises. On the other hand, no plateaus occur with natural gradient descent regardless of the correlation for the limit of a low learning rate. Analytical results support these dynamics around the saddle point.

収録刊行物

被引用文献 (6)*注記

もっと見る

参考文献 (9)*注記

もっと見る

キーワード

詳細情報 詳細情報について

問題の指摘

ページトップへ