arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1709.04054stat.MLcs.LGcs.NE
收藏

用双极激活函数将平均激活移向零

Shifting Mean Activation Towards Zero with Bipolar Activation Functions

Lars Eidnes, Arild Nøkland

摘要:

我们提出了一个简单的扩展ReLU家族的激活函数,使他们能够移动的平均激活跨层向零。结合适当的权重初始化,这就减少了对归一化层的需求。我们探索了多达144层的深度香草递归神经网络(RNN)的训练,并表明双极激活函数有助于在这种情况下学习。在Penn Treebank和Text 8语言建模任务中,我们获得了有竞争力的结果,改进了非门控网络的最佳报告结果。在没有批量标准化的卷积神经网络的实验中,我们发现双极激活会更快地降低训练误差,并导致CIFAR-10分类任务的测试误差更低。

英文摘要:

We propose a simple extension to the ReLU-family of activation functions that allows them to shift the mean activation across a layer towards zero. Combined with proper weight initialization, this alleviates the need for normalization layers. We explore the training of deep vanilla recurrent neural networks (RNNs) with up to 144 layers, and show that bipolar activation functions help learning in this setting. On the Penn Treebank and Text8 language modeling tasks we obtain competitive results, improving on the best reported results for non-gated networks. In experiments with convolutional neural networks without batch normalization, we find that bipolar activations produce a faster drop in training error, and results in a lower test error on the CIFAR-10 classification 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。

↑