arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2001.06370cs.LGcs.PFstat.ML
收藏

近似激活函数

Approximating Activation Functions

University of Cambridge(剑桥大学)

Nicholas Gerard Timmons, Andrew Rice

摘要:

ReLU被广泛认为是神经网络中激活函数的默认选择。然而,存在需要更复杂的功能的情况。特别是,递归神经网络(如LSTM)广泛使用双曲正切函数和sigmoid函数。这些函数的计算成本很高。我们使用函数近似技术来开发这些函数的替代品,并在三种流行的网络配置上对其进行经验评估。我们找到了安全的近似值,可以在CPU上的训练时间上提高10%到37%。这些近似适用于我们考虑的所有情况,我们相信是使用这些激活函数的所有网络的适当替代。我们还开发了范围近似,仅适用于在某些情况下,由于其输入域的限制。我们的范围近似在网络训练时间上提高了20%到53%的性能。我们的函数也匹配或大大超过Theano和Word 2 Vec实现中使用的ad-hoc近似。

英文摘要:

ReLU is widely seen as the default choice for activation functions in neural networks. However, there are cases where more complicated functions are required. In particular, recurrent neural networks (such as LSTMs) make extensive use of both hyperbolic tangent and sigmoid functions. These functions are expensive to compute. We used function approximation techniques to develop replacements for these functions and evaluated them empirically on three popular network configurations. We find safe approximations that yield a 10% to 37% improvement in training times on the CPU. These approximations were suitable for all cases we considered and we believe are appropriate replacements for all networks using these activation functions. We also develop ranged approximations which only apply in some cases due to restrictions on their input domain. Our ranged approximations yield a performance improvement of 20% to 53% in network training time. Our functions also match or considerably out perform the ad-hoc approximations used in Theano and the implementation of Word2Vec.

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。

↑