Recurrent Neural Networks with Flexible Gates using Kernel Activation Functions
This work offers an incremental improvement for researchers and practitioners in machine learning by enhancing gated RNNs for sequential data analysis.
The paper tackled the limitation of standard sigmoid gates in recurrent neural networks by proposing a more flexible gate architecture using kernel activation functions, which improved accuracy on sequential MNIST variants with negligible computational cost and faster training.
Gated recurrent neural networks have achieved remarkable results in the analysis of sequential data. Inside these networks, gates are used to control the flow of information, allowing to model even very long-term dependencies in the data. In this paper, we investigate whether the original gate equation (a linear projection followed by an element-wise sigmoid) can be improved. In particular, we design a more flexible architecture, with a small number of adaptable parameters, which is able to model a wider range of gating functions than the classical one. To this end, we replace the sigmoid function in the standard gate with a non-parametric formulation extending the recently proposed kernel activation function (KAF), with the addition of a residual skip-connection. A set of experiments on sequential variants of the MNIST dataset shows that the adoption of this novel gate allows to improve accuracy with a negligible cost in terms of computational power and with a large speed-up in the number of training iterations.