arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1512.05830cs.CVcs.LG
收藏

基于中继反向传播的深度卷积神经网络有效学习

Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networks

University of Chinese Academy of Sciences(中国科学院大学);Peking University(北京大学)

Li Shen, Zhouchen Lin, Qingming Huang

摘要:

近年来,对卷积神经网络进行深度学习成为一种趋势。然而,许多经验证据表明,性能改善不能简单地通过堆叠更多的层。在本文中,我们认为这个问题从信息理论的角度来看,并提出了一种新的方法中继反向传播,鼓励传播的有效信息通过网络在训练阶段。该方法在ILSVRC 2015场景分类挑战赛中获得第一名。在两个具有挑战性的大规模数据集上的大量实验表明,我们的方法的有效性不限于特定的数据集或网络架构。我们的模型稍后将提供给研究界。

英文摘要:

Learning deeper convolutional neural networks becomes a tendency in recent years. However, many empirical evidences suggest that performance improvement cannot be gained by simply stacking more layers. In this paper, we consider the issue from an information theoretical perspective, and propose a novel method Relay Backpropagation, that encourages the propagation of effective information through the network in training stage. By virtue of the method, we achieved the first place in ILSVRC 2015 Scene Classification Challenge. Extensive experiments on two challenging large scale datasets demonstrate the effectiveness of our method is not restricted to a specific dataset or network architecture. Our models will be available to the research community later.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑