2016/04/11 by Yuan Gao, Dorota Glowacka, Gao, Yuan +2 · 1 voice · 11 citations
Computer Science · Engineering · Mathematics · #Algorithm #Artificial intelligence #Artificial neural network #Computation #Computer science #Deep learning #Dependency (UML) #Ferroelectric and Negative Capacitance Devices #Graph #Information flow #Logic gate #Long short term memory #Mathematics #Multiplicative function #Neural Networks and Applications #Recurrent neural network #Sequence (biology) #Simple (philosophy) #Stochastic Gradient Optimization Techniques #Theoretical computer science #cs.NE
paper · pdf · doi:10.48550/arxiv.1604.02910
published in arXiv (Cornell University), 350-365 (Cornell University) · This paper has been withdrawn by the author due to lacking of enough experiments
openalex publication_date 2016/04/11 · arxiv created 2016/05/13 · arxiv updated 2016/05/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01
This paper introduces two recurrent neural network structures called Simple Gated Unit (SGU) and Deep Simple Gated Unit (DSGU), which are general structures for learning long term dependencies. Compared to traditional Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), both structures require fewer parameters and less computation time in sequence classification tasks. Unlike GRU and LSTM, which require more than one gates to control information flow in the network, SGU and DSGU only use one multiplicative gate to control the flow of information. We show that this difference can accelerate the learning speed in tasks that require long dependency information. We also show that DSGU is more numerically stable than SGU. In addition, we also propose a standard way of representing inner structure of RNN called RNN Conventional Graph (RCG), which helps analyzing the relationship between input units and hidden units of RNN.