Abstract
Recurrent neural networks (RNNs) have enjoyed
great success in speech recognition, natural language processing, etc. Many variants of RNNs
have been proposed, including vanilla RNNs,
LSTMs, and GRUs. However, current architectures
are not particularly adept at dealing with tasks involving multi-faceted contents, i.e., data with a bimodal or multimodal distribution. In this work, we
solve this problem by proposing Multiple-Weight
RNNs and LSTMs, which rely on multiple weight
matrices to better mimic the human ability of
switching between contexts. We present a framework for adapting RNN-based models and analyze
the properties of this approach. Our detailed experimental results show that our model outperforms
previous work across a range of different tasks and
datasets.