arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1602.08210cs.LGcs.NE
收藏

循环神经网络的结构复杂性度量

Architectural Complexity Measures of Recurrent Neural Networks

MILA, Université de Montréal(蒙特利尔大学 MILA);University of Toronto(多伦多大学);Institut des Hautes Études Scientifiques, France(法国高等科学研究所);Carnegie Mellon University(卡内基梅隆大学)

Saizheng Zhang, Yuhuai Wu, Tong Che, Zhouhan Lin, Roland Memisevic, Ruslan Salakhutdinov, Yoshua Bengio

摘要:

本文系统地分析了循环神经网络(RNN)的连接结构。我们的主要贡献是两方面的:首先,我们提出了一个严格的图论框架,描述了一般RNN的连接架构。其次,我们提出了RNN的三种架构复杂度度量:(a)递归深度,它捕获RNN随时间的非线性复杂度,(b)前馈深度,它捕获局部输入输出非线性(类似于前馈神经网络(FNN)中的“深度”),以及(c)递归跳跃系数,它捕获信息随时间传播的速度。我们严格证明了每一个措施的存在性和可计算性。我们的实验结果表明,RNN可能受益于更大的递归深度和前馈深度。我们进一步证明,增加经常性跳跃系数提供了长期依赖问题的性能提升。

英文摘要:

In this paper, we systematically analyze the connecting architectures of recurrent neural networks (RNNs). Our main contribution is twofold: first, we present a rigorous graph-theoretic framework describing the connecting architectures of RNNs in general. Second, we propose three architecture complexity measures of RNNs: (a) the recurrent depth, which captures the RNN's over-time nonlinear complexity, (b) the feedforward depth, which captures the local input-output nonlinearity (similar to the "depth" in feedforward neural networks (FNNs)), and (c) the recurrent skip coefficient which captures how rapidly the information propagates over time. We rigorously prove each measure's existence and computability. Our experimental results show that RNNs might benefit from larger recurrent depth and feedforward depth. We further demonstrate that increasing recurrent skip coefficient offers performance boosts on long term dependency problems.

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。

↑