MultiscaleRNN

This is a PyTorch implementation of the ICLR 2017 paper "HIERARCHICAL MULTISCALE RECURRENT NEURAL NETWORKS" (https://openreview.net/pdf?id=S1di0sfgl).

Acknowledgement

I have no affiliation with the authors of the paper and I am implementing this code for non-commercial reasons.

The authors published their official Tensorflow implementation here (https://github.com/n-s-f/hierarchical-rnn), so check it out for something that is guaranteed to work as intended. From what I gather, their implementation is slightly different than mine, so that may be something I will investigate in the future.

You should also cite the paper if you use any of this code for your research:

@article{DBLP:journals/corr/ChungAB16,
  author    = {Junyoung Chung and
               Sungjin Ahn and
               Yoshua Bengio},
  title     = {Hierarchical Multiscale Recurrent Neural Networks},
  journal   = {CoRR},
  volume    = {abs/1609.01704},
  year      = {2016}
}