/counting_dimensions

demonstration for our ACL 2018 paper, "On the Practical Computational Power of Finite Precision RNNs for Language Recognition"

Primary LanguageJupyter Notebook

ACL support repository

This is a simple repository to host files for the following google colab notebook: https://drive.google.com/file/d/1Ya4Dhih8JU-do9t4gcxkU4pgb9Gm8dmo/view?usp=sharing (though you can also just clone the repository and run the notebook in it on your own computer :) )

The notebook is a demonstration of the results of our paper, "On the Practical Computational Power of Finite Precision RNNs for Language Recognition". It holds 4 pretrained networks: the languages $a^nb^n$ and $a^nb^nc^n$, on LSTM and GRU networks. You can also train your own networks.

Citation

You can cite this work using:

@InProceedings{P18-2117, author = "Weiss, Gail and Goldberg, Yoav and Yahav, Eran", title = "On the Practical Computational Power of Finite Precision RNNs for Language Recognition", booktitle = "Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)", year = "2018", publisher = "Association for Computational Linguistics", pages = "740--745", location = "Melbourne, Australia", url = "http://aclweb.org/anthology/P18-2117" }