标签：
Google DeepMind又一神作
Projected Natural Gradient Descent algorithm (PRONG) better than SGD as evidenced by the boost in performance offered by batch normalization (BN)
版权声明：本文为博主原创文章，未经博主允许不得转载。
[Fri 26 Jun 2015 ~ Thu 2 Jul 2015] Deep Learning in arxiv
标签：
原文地址：http://blog.csdn.net/sunbaigui/article/details/46802121
