arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1705.02414cs.LGstat.ML
收藏

MXNet中循环神经网络批量构造策略的综合研究

A comprehensive study of batch construction strategies for recurrent neural networks in MXNet

AppTek(AppTek 公司)

Patrick Doetsch, Pavel Golik, Hermann Ney

摘要:

在这项工作中,我们比较了不同的小批量训练的循环神经网络批量建设方法。虽然像TensorFlow和MXNet这样的流行实现建议采用分桶方法来提高循环训练过程的并行化能力,但我们提出了一种简单的排序策略,以随机交替排序的方式排列训练序列。我们比较我们的方法序列桶以及其他各种批量建设策略的CHiME-4嘈杂的语音识别语料库。实验表明,我们的交替排序方法能够在训练时间和识别性能方面进行竞争,同时在概念上更容易实现。

英文摘要:

In this work we compare different batch construction methods for mini-batch training of recurrent neural networks. While popular implementations like TensorFlow and MXNet suggest a bucketing approach to improve the parallelization capabilities of the recurrent training process, we propose a simple ordering strategy that arranges the training sequences in a stochastic alternatingly sorted way. We compare our method to sequence bucketing as well as various other batch construction strategies on the CHiME-4 noisy speech recognition corpus. The experiments show that our alternated sorting approach is able to compete both in training time and recognition performance while being conceptually simpler to implement.

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。

↑