<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="irie-etal-2023-practical">
    <titleInfo>
        <title>Practical Computational Power of Linear Transformers and Their Recurrent and Self-Referential Extensions</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Kazuki</namePart>
        <namePart type="family">Irie</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Róbert</namePart>
        <namePart type="family">Csordás</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jürgen</namePart>
        <namePart type="family">Schmidhuber</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2023-12</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Houda</namePart>
            <namePart type="family">Bouamor</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Juan</namePart>
            <namePart type="family">Pino</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Kalika</namePart>
            <namePart type="family">Bali</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Singapore</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Recent studies of the computational power of recurrent neural networks (RNNs) reveal a hierarchy of RNN architectures, given real-time and finite-precision assumptions. Here we study auto-regressive Transformers with linearised attention, a.k.a. linear Transformers (LTs) or Fast Weight Programmers (FWPs). LTs are special in the sense that they are equivalent to RNN-like sequence processors with a fixed-size state, while they can also be expressed as the now-popular self-attention networks. We show that many well-known results for the standard Transformer directly transfer to LTs/FWPs. Our formal language recognition experiments demonstrate how recently proposed FWP extensions such as recurrent FWPs and self-referential weight matrices successfully overcome certain limitations of the LT, e.g., allowing for generalisation on the parity problem. Our code is public.</abstract>
    <identifier type="citekey">irie-etal-2023-practical</identifier>
    <identifier type="doi">10.18653/v1/2023.emnlp-main.588</identifier>
    <location>
        <url>https://aclanthology.org/2023.emnlp-main.588</url>
    </location>
    <part>
        <date>2023-12</date>
        <extent unit="page">
            <start>9455</start>
            <end>9465</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Practical Computational Power of Linear Transformers and Their Recurrent and Self-Referential Extensions
%A Irie, Kazuki
%A Csordás, Róbert
%A Schmidhuber, Jürgen
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F irie-etal-2023-practical
%X Recent studies of the computational power of recurrent neural networks (RNNs) reveal a hierarchy of RNN architectures, given real-time and finite-precision assumptions. Here we study auto-regressive Transformers with linearised attention, a.k.a. linear Transformers (LTs) or Fast Weight Programmers (FWPs). LTs are special in the sense that they are equivalent to RNN-like sequence processors with a fixed-size state, while they can also be expressed as the now-popular self-attention networks. We show that many well-known results for the standard Transformer directly transfer to LTs/FWPs. Our formal language recognition experiments demonstrate how recently proposed FWP extensions such as recurrent FWPs and self-referential weight matrices successfully overcome certain limitations of the LT, e.g., allowing for generalisation on the parity problem. Our code is public.
%R 10.18653/v1/2023.emnlp-main.588
%U https://aclanthology.org/2023.emnlp-main.588
%U https://doi.org/10.18653/v1/2023.emnlp-main.588
%P 9455-9465

Markdown (Informal)

[Practical Computational Power of Linear Transformers and Their Recurrent and Self-Referential Extensions](https://aclanthology.org/2023.emnlp-main.588) (Irie et al., EMNLP 2023)

• Practical Computational Power of Linear Transformers and Their Recurrent and Self-Referential Extensions (Irie et al., EMNLP 2023)

ACL

• Kazuki Irie, Róbert Csordás, and Jürgen Schmidhuber. 2023. Practical Computational Power of Linear Transformers and Their Recurrent and Self-Referential Extensions. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 9455–9465, Singapore. Association for Computational Linguistics.