Adaptive convolution for text classification

Byung Ju Choi; Jun Hyung Park; Sang Keun Lee

Adaptive convolution for text classification

Byung Ju Choi, Jun Hyung Park, Sang Keun Lee

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper, we present an adaptive convolution for text classification to give stronger flexibility to convolutional neural networks (CNNs). Unlike traditional convolutions that use the same set of filters regardless of different inputs, the adaptive convolution employs adaptively generated convolutional filters that are conditioned on inputs. We achieve this by attaching filter-generating networks, which are carefully designed to generate input-specific filters, to convolution blocks in existing CNNs. We show the efficacy of our approach in existing CNNs based on our performance evaluation. Our evaluation indicates that adaptive convolutions improve all the baselines, without any exception, as much as up to 2.6 percentage point in seven benchmark text classification datasets.

Original language	English
Title of host publication	Long and Short Papers
Publisher	Association for Computational Linguistics (ACL)
Pages	2475-2485
Number of pages	11
ISBN (Electronic)	9781950737130
Publication status	Published - 2019
Event	2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2019 - Minneapolis, United States Duration: 2019 Jun 2 → 2019 Jun 7

Publication series

Name	NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference
Volume	1

Conference

Conference	2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2019
Country	United States
City	Minneapolis
Period	19/6/2 → 19/6/7

ASJC Scopus subject areas

Language and Linguistics
Computer Science Applications
Linguistics and Language

Fingerprint Dive into the research topics of 'Adaptive convolution for text classification'. Together they form a unique fingerprint.

Cite this

Adaptive convolution for text classification. / Choi, Byung Ju; Park, Jun Hyung; Lee, Sang Keun.

Long and Short Papers. Association for Computational Linguistics (ACL), 2019. p. 2475-2485 (NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference; Vol. 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Choi, BJ, Park, JH & Lee, SK 2019, Adaptive convolution for text classification. in Long and Short Papers. NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference, vol. 1, Association for Computational Linguistics (ACL), pp. 2475-2485, 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2019, Minneapolis, United States, 19/6/2.

@inproceedings{5afb68a3e2db420ea68f1605b45856fc,

title = "Adaptive convolution for text classification",

abstract = "In this paper, we present an adaptive convolution for text classification to give stronger flexibility to convolutional neural networks (CNNs). Unlike traditional convolutions that use the same set of filters regardless of different inputs, the adaptive convolution employs adaptively generated convolutional filters that are conditioned on inputs. We achieve this by attaching filter-generating networks, which are carefully designed to generate input-specific filters, to convolution blocks in existing CNNs. We show the efficacy of our approach in existing CNNs based on our performance evaluation. Our evaluation indicates that adaptive convolutions improve all the baselines, without any exception, as much as up to 2.6 percentage point in seven benchmark text classification datasets.",

author = "Choi, {Byung Ju} and Park, {Jun Hyung} and Lee, {Sang Keun}",

year = "2019",

language = "English",

series = "NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference",

publisher = "Association for Computational Linguistics (ACL)",

pages = "2475--2485",

booktitle = "Long and Short Papers",

note = "2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2019 ; Conference date: 02-06-2019 Through 07-06-2019",

}

TY - GEN

T1 - Adaptive convolution for text classification

AU - Choi, Byung Ju

AU - Park, Jun Hyung

AU - Lee, Sang Keun

PY - 2019

Y1 - 2019

N2 - In this paper, we present an adaptive convolution for text classification to give stronger flexibility to convolutional neural networks (CNNs). Unlike traditional convolutions that use the same set of filters regardless of different inputs, the adaptive convolution employs adaptively generated convolutional filters that are conditioned on inputs. We achieve this by attaching filter-generating networks, which are carefully designed to generate input-specific filters, to convolution blocks in existing CNNs. We show the efficacy of our approach in existing CNNs based on our performance evaluation. Our evaluation indicates that adaptive convolutions improve all the baselines, without any exception, as much as up to 2.6 percentage point in seven benchmark text classification datasets.

AB - In this paper, we present an adaptive convolution for text classification to give stronger flexibility to convolutional neural networks (CNNs). Unlike traditional convolutions that use the same set of filters regardless of different inputs, the adaptive convolution employs adaptively generated convolutional filters that are conditioned on inputs. We achieve this by attaching filter-generating networks, which are carefully designed to generate input-specific filters, to convolution blocks in existing CNNs. We show the efficacy of our approach in existing CNNs based on our performance evaluation. Our evaluation indicates that adaptive convolutions improve all the baselines, without any exception, as much as up to 2.6 percentage point in seven benchmark text classification datasets.

UR - http://www.scopus.com/inward/record.url?scp=85085544480&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85085544480&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85085544480

T3 - NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference

SP - 2475

EP - 2485

BT - Long and Short Papers

PB - Association for Computational Linguistics (ACL)

T2 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2019

Y2 - 2 June 2019 through 7 June 2019

ER -