Dynamics of Learning In Hierarchical Models - Singularity and Milnor Attractor

被引：2

作者：

Amari, Shun-ichi ^{[1
]}

Ozeki, Tomoko ^{[1
]}

Cousseau, Florent ^{[1
]}

Wei, Haikun ^{[1
]}

机构：

[1] RIKEN, Brain Sci Inst, Saitama 3510198, Japan

来源：

ADVANCES IN COGNITIVE NEURODYNAMICS (II) | 2011年

关键词：

Dynamics of learning; On-line learning; Multilayer-perceptron; Singularity; Milnor attractor; MULTILAYER PERCEPTRONS; NETWORKS;

D O I：

10.1007/978-90-481-9695-1_1

中图分类号：

Q189 [神经科学];

学科分类号：

071006 ;

摘要：

We study the dynamics of learning in a hierarchical model such as multilayer perceptron. Such a model includes singularities, which affects its dynamics seriously. The Milnor type attractors appear, because of the singularity. We will show its trajectories explicitly, and present the topological nature of the singularities.

引用

页码：3 / 9

页数：7

共 7 条

[1] Natural gradient works efficiently in learning [J].