Minimum Description Length Recurrent Neural Networks
Computation and Language
2022-04-01 v4
Abstract
We train neural networks to optimize a Minimum Description Length score, i.e., to balance between the complexity of the network and its accuracy at a task. We show that networks optimizing this objective function master tasks involving memory challenges and go beyond context-free languages. These learners master languages such as , , , , and they perform addition. Moreover, they often do so with 100% accuracy. The networks are small, and their inner workings are transparent. We thus provide formal proofs that their perfect accuracy holds not only on a given test set, but for any input sequence. To our knowledge, no other connectionist model has been shown to capture the underlying grammars for these languages in full generality.
Cite
@article{arxiv.2111.00600,
title = {Minimum Description Length Recurrent Neural Networks},
author = {Nur Lan and Michal Geyer and Emmanuel Chemla and Roni Katzir},
journal= {arXiv preprint arXiv:2111.00600},
year = {2022}
}
Comments
15 pages