papers AI Learner
The Github is limit! Click to go to the new site.

Dynamic Cell Structure via Recursive-Recurrent Neural Networks

2019-05-25
Xin Qian, Matthew Kennedy, Diego Klabjan

Abstract

In a recurrent setting, conventional approaches to neural architecture search find and fix a general model for all data samples and time steps. We propose a novel algorithm that can dynamically search for the structure of cells in a recurrent neural network model. Based on a combination of recurrent and recursive neural networks, our algorithm is able to construct customized cell structures for each data sample and time step, allowing for a more efficient architecture search than existing models. Experiments on three common datasets show that the algorithm discovers high-performance cell architectures and achieves better prediction accuracy compared to the GRU structure for language modelling and sentiment analysis.

Abstract (translated by Google)
URL

https://arxiv.org/abs/1905.10540

PDF

https://arxiv.org/pdf/1905.10540


Similar Posts

Comments