学习的均场理论:从动力学到统计
无序系统与神经网络
2007-05-23 v1
摘要
我们利用腔方法和图谱方法,构建了受限示例集批量学习动力学的模型。对 Green's 函数和腔激活分布的仿真结果良好地支持了该理论。学习动力学趋于与静态腔方法一致的稳态。我们审阅了粗糙能量景观的图景。
关键词
引用
@article{arxiv.cond-mat/0006251,
title = {Mean-field theory of learning: from dynamics to statics},
author = {K. Y. Michael Wong and S. Li and Peixun Luo},
journal= {arXiv preprint arXiv:cond-mat/0006251},
year = {2007}
}
备注
13 pages, 5 figures, to appear in "Advanced Mean Field Methods - Theory and Practice", edited by M. Opper and D. Saad, MIT Press