TanhSoft -- a family of activation functions combining Tanh and Softplus
arXiv:2009.03863
Abstract
Deep learning at its core, contains functions that are composition of a linear transformation with a non-linear function known as activation function. In past few years, there is an increasing interest in construction of novel activation functions resulting in better learning. In this work, we propose a family of novel activation functions, namely TanhSoft, with four undetermined hyper-parameters of the form tanh(αx+βe^{γx})ln(δ+e^x) and tune these hyper-parameters to obtain activation functions which are shown to outperform several well known activation functions. For instance, replacing ReLU with xtanh(0.6e^x)improves top-1 classification accuracy on CIFAR-10 by 0.46% for DenseNet-169 and 0.7% for Inception-v3 while with tanh(0.87x)ln(1 +e^x) top-1 classification accuracy on CIFAR-100 improves by 1.24% for DenseNet-169 and 2.57% for SimpleNet model.
arXiv comment11 pages
References in corpus (10)
- MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
- Empirical Evaluation of Rectified Activations in Convolutional Network
- Activation Functions: Comparison of trends in Practice and Research for Deep Learning
- Searching for Activation Functions
- Deep Voice 3: Scaling Text-to-Speech with Convolutional Sequence Learning
- Lets keep it simple, Using simple architectures to outperform deeper and more complex architectures
- Improving Deep Learning by Inverse Square Root Linear Units (ISRLUs)
- Compositional Distributional Semantics with Long Short Term Memory
- E-swish: Adjusting Activations to Different Network Depths