Minimal Gated Unit for Recurrent Neural Networks
This work addresses the difficulty in understanding and optimizing RNNs for sequence data applications, though it is incremental as it builds on existing gated units like LSTM and GRU.
The authors tackled the complexity of recurrent neural networks by proposing the Minimal Gated Unit (MGU), a gated hidden unit with only one gate, which achieves comparable accuracy to GRU while offering a simpler structure, fewer parameters, and faster training.
Recently recurrent neural networks (RNN) has been very successful in handling sequence data. However, understanding RNN and finding the best practices for RNN is a difficult task, partly because there are many competing and complex hidden units (such as LSTM and GRU). We propose a gated unit for RNN, named as Minimal Gated Unit (MGU), since it only contains one gate, which is a minimal design among all gated hidden units. The design of MGU benefits from evaluation results on LSTM and GRU in the literature. Experiments on various sequence data show that MGU has comparable accuracy with GRU, but has a simpler structure, fewer parameters, and faster training. Hence, MGU is suitable in RNN's applications. Its simple architecture also means that it is easier to evaluate and tune, and in principle it is easier to study MGU's properties theoretically and empirically.