The unreasonable effectiveness of the forget gate
This work addresses the need for more efficient and effective recurrent neural networks for machine learning practitioners, though it is incremental as it builds on existing LSTM research.
The paper tackled the problem of simplifying LSTM networks by questioning the necessity of all gates, showing that a forget-gate-only version with chrono-initialized biases outperforms the standard LSTM on benchmarks like MNIST (99% vs. 98.5%) and pMNIST (92.5% vs. 91%).
Given the success of the gated recurrent unit, a natural question is whether all the gates of the long short-term memory (LSTM) network are necessary. Previous research has shown that the forget gate is one of the most important gates in the LSTM. Here we show that a forget-gate-only version of the LSTM with chrono-initialized biases, not only provides computational savings but outperforms the standard LSTM on multiple benchmark datasets and competes with some of the best contemporary models. Our proposed network, the JANET, achieves accuracies of 99% and 92.5% on the MNIST and pMNIST datasets, outperforming the standard LSTM which yields accuracies of 98.5% and 91%.