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 1508.02774cs.NE
收藏

LSTM网络的基准测试

Benchmarking of LSTM Networks

Thomas M. Breuel

摘要:

LSTM(长短期记忆)递归神经网络在许多应用领域都非常成功。本技术报告描述了使用MNIST和UW3数据库对LSTM网络进行基准测试,并探讨了不同架构和超参数选择对性能的影响。重要的调查结果包括:(1)LSTM的性能平稳地依赖于学习率,(2)迭代和动量对性能没有显著影响,(3)softmax训练优于最小二乘训练,(4)窥视孔单元没有用,(5)标准非线性(tanh和sigmoid)表现最好,(6)双向训练结合CTC比其他方法表现更好。

英文摘要:

LSTM (Long Short-Term Memory) recurrent neural networks have been highly successful in a number of application areas. This technical report describes the use of the MNIST and UW3 databases for benchmarking LSTM networks and explores the effect of different architectural and hyperparameter choices on performance. Significant findings include: (1) LSTM performance depends smoothly on learning rates, (2) batching and momentum has no significant effect on performance, (3) softmax training outperforms least square training, (4) peephole units are not useful, (5) the standard non-linearities (tanh and sigmoid) perform best, (6) bidirectional training combined with CTC performs better than other methods.

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。

↑