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 1901.02671cs.CLcs.LGcs.NE
收藏

是时候去Swish了吗?跨NLP任务比较深度学习激活函数

Is it Time to Swish? Comparing Deep Learning Activation Functions Across NLP tasks

Ubiquitous Knowledge Processing Lab (UKP-TUDA)(普适知识处理实验室(UKP-TUDA));Department of Computer Science(计算机科学系);Technische Universität Darmstadt(达姆施塔特工业大学)

Steffen Eger, Paul Youssef, Iryna Gurevych

摘要:

激活函数在神经网络中起着至关重要的作用,因为它们是归因于深度学习成功故事的非线性。目前最流行的激活函数之一是ReLU,但最近已经提出或“发现”了几个竞争对手,包括LReLU函数和swish。虽然大多数作品在少数任务(通常来自图像分类)上比较新提出的激活函数,并与少数竞争对手(通常是ReLU)进行比较,但我们在8个不同的NLP任务中对21个激活函数进行了首次大规模比较。我们发现,一个很大程度上未知的激活函数执行最稳定的所有任务,所谓的惩罚双曲正切函数。我们还表明,它可以成功地取代LSTM单元中的sigmoid和tanh门,在具有挑战性的NLP任务中,比标准选择提高了2个百分点。

英文摘要:

Activation functions play a crucial role in neural networks because they are the nonlinearities which have been attributed to the success story of deep learning. One of the currently most popular activation functions is ReLU, but several competitors have recently been proposed or 'discovered', including LReLU functions and swish. While most works compare newly proposed activation functions on few tasks (usually from image classification) and against few competitors (usually ReLU), we perform the first large-scale comparison of 21 activation functions across eight different NLP tasks. We find that a largely unknown activation function performs most stably across all tasks, the so-called penalized tanh function. We also show that it can successfully replace the sigmoid and tanh gates in LSTM cells, leading to a 2 percentage point (pp) improvement over the standard choices on a challenging NLP task.

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。

↑