"Despite lacking trainable recurrent layers, stacked QRNNs have better predictive accuracy than stacked LSTMs of the same hidden size. Due to their increased parallelism, they are up to 16 times faster at train and test time"
RNN is currently a 30th of the speed of the CNN per transaction. We have two updates on the horizon, one to double performance and then another to quadruple it. This may make those updates unnecessary if easy to implement.
"Despite lacking trainable recurrent layers, stacked QRNNs have better predictive accuracy than stacked LSTMs of the same hidden size. Due to their increased parallelism, they are up to 16 times faster at train and test time"
https://arxiv.org/abs/1611.01576
RNN is currently a 30th of the speed of the CNN per transaction. We have two updates on the horizon, one to double performance and then another to quadruple it. This may make those updates unnecessary if easy to implement.