Dimension Reduction Approach for Interpretability of Sequence to Sequence Recurrent Neural Networks

Open in new window