@article{lan-etal-2022-minimum,
title = "Minimum Description Length Recurrent Neural Networks",
author = "Lan, Nur and
Geyer, Michal and
Chemla, Emmanuel and
Katzir, Roni",
editor = "Roark, Brian and
Nenkova, Ani",
journal = "Transactions of the Association for Computational Linguistics",
volume = "10",
year = "2022",
address = "Cambridge, MA",
publisher = "MIT Press",
url = "https://aclanthology.org/2022.tacl-1.45/",
doi = "10.1162/tacl_a_00489",
pages = "785--799",
abstract = "We train neural networks to optimize a Minimum Description Length score, that is, 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 anbn, anbncn, anb2n, anbmcn +m, 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."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lan-etal-2022-minimum">
<titleInfo>
<title>Minimum Description Length Recurrent Neural Networks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nur</namePart>
<namePart type="family">Lan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michal</namePart>
<namePart type="family">Geyer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Emmanuel</namePart>
<namePart type="family">Chemla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roni</namePart>
<namePart type="family">Katzir</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<genre authority="bibutilsgt">journal article</genre>
<relatedItem type="host">
<titleInfo>
<title>Transactions of the Association for Computational Linguistics</title>
</titleInfo>
<originInfo>
<issuance>continuing</issuance>
<publisher>MIT Press</publisher>
<place>
<placeTerm type="text">Cambridge, MA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">periodical</genre>
<genre authority="bibutilsgt">academic journal</genre>
</relatedItem>
<abstract>We train neural networks to optimize a Minimum Description Length score, that is, 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 anbn, anbncn, anb2n, anbmcn +m, 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.</abstract>
<identifier type="citekey">lan-etal-2022-minimum</identifier>
<identifier type="doi">10.1162/tacl_a_00489</identifier>
<location>
<url>https://aclanthology.org/2022.tacl-1.45/</url>
</location>
<part>
<date>2022</date>
<detail type="volume"><number>10</number></detail>
<extent unit="page">
<start>785</start>
<end>799</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Journal Article
%T Minimum Description Length Recurrent Neural Networks
%A Lan, Nur
%A Geyer, Michal
%A Chemla, Emmanuel
%A Katzir, Roni
%J Transactions of the Association for Computational Linguistics
%D 2022
%V 10
%I MIT Press
%C Cambridge, MA
%F lan-etal-2022-minimum
%X We train neural networks to optimize a Minimum Description Length score, that is, 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 anbn, anbncn, anb2n, anbmcn +m, 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.
%R 10.1162/tacl_a_00489
%U https://aclanthology.org/2022.tacl-1.45/
%U https://doi.org/10.1162/tacl_a_00489
%P 785-799
Markdown (Informal)
[Minimum Description Length Recurrent Neural Networks](https://aclanthology.org/2022.tacl-1.45/) (Lan et al., TACL 2022)
ACL