@inproceedings{gupta-etal-2020-effective,
title = "Effective Few-Shot Classification with Transfer Learning",
author = "Gupta, Aakriti and
Thadani, Kapil and
O{'}Hare, Neil",
editor = "Scott, Donia and
Bel, Nuria and
Zong, Chengqing",
booktitle = "Proceedings of the 28th International Conference on Computational Linguistics",
month = dec,
year = "2020",
address = "Barcelona, Spain (Online)",
publisher = "International Committee on Computational Linguistics",
url = "https://aclanthology.org/2020.coling-main.92",
doi = "10.18653/v1/2020.coling-main.92",
pages = "1061--1066",
abstract = "Few-shot learning addresses the the problem of learning based on a small amount of training data. Although more well-studied in the domain of computer vision, recent work has adapted the Amazon Review Sentiment Classification (ARSC) text dataset for use in the few-shot setting. In this work, we use the ARSC dataset to study a simple application of transfer learning approaches to few-shot classification. We train a single binary classifier to learn all few-shot classes jointly by prefixing class identifiers to the input text. Given the text and class, the model then makes a binary prediction for that text/class pair. Our results show that this simple approach can outperform most published results on this dataset. Surprisingly, we also show that including domain information as part of the task definition only leads to a modest improvement in model accuracy, and zero-shot classification, without further fine-tuning on few-shot domains, performs equivalently to few-shot classification. These results suggest that the classes in the ARSC few-shot task, which are defined by the intersection of domain and rating, are actually very similar to each other, and that a more suitable dataset is needed for the study of few-shot text classification.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gupta-etal-2020-effective">
<titleInfo>
<title>Effective Few-Shot Classification with Transfer Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aakriti</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kapil</namePart>
<namePart type="family">Thadani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Neil</namePart>
<namePart type="family">O’Hare</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 28th International Conference on Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Donia</namePart>
<namePart type="family">Scott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nuria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chengqing</namePart>
<namePart type="family">Zong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>International Committee on Computational Linguistics</publisher>
<place>
<placeTerm type="text">Barcelona, Spain (Online)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Few-shot learning addresses the the problem of learning based on a small amount of training data. Although more well-studied in the domain of computer vision, recent work has adapted the Amazon Review Sentiment Classification (ARSC) text dataset for use in the few-shot setting. In this work, we use the ARSC dataset to study a simple application of transfer learning approaches to few-shot classification. We train a single binary classifier to learn all few-shot classes jointly by prefixing class identifiers to the input text. Given the text and class, the model then makes a binary prediction for that text/class pair. Our results show that this simple approach can outperform most published results on this dataset. Surprisingly, we also show that including domain information as part of the task definition only leads to a modest improvement in model accuracy, and zero-shot classification, without further fine-tuning on few-shot domains, performs equivalently to few-shot classification. These results suggest that the classes in the ARSC few-shot task, which are defined by the intersection of domain and rating, are actually very similar to each other, and that a more suitable dataset is needed for the study of few-shot text classification.</abstract>
<identifier type="citekey">gupta-etal-2020-effective</identifier>
<identifier type="doi">10.18653/v1/2020.coling-main.92</identifier>
<location>
<url>https://aclanthology.org/2020.coling-main.92</url>
</location>
<part>
<date>2020-12</date>
<extent unit="page">
<start>1061</start>
<end>1066</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Effective Few-Shot Classification with Transfer Learning
%A Gupta, Aakriti
%A Thadani, Kapil
%A O’Hare, Neil
%Y Scott, Donia
%Y Bel, Nuria
%Y Zong, Chengqing
%S Proceedings of the 28th International Conference on Computational Linguistics
%D 2020
%8 December
%I International Committee on Computational Linguistics
%C Barcelona, Spain (Online)
%F gupta-etal-2020-effective
%X Few-shot learning addresses the the problem of learning based on a small amount of training data. Although more well-studied in the domain of computer vision, recent work has adapted the Amazon Review Sentiment Classification (ARSC) text dataset for use in the few-shot setting. In this work, we use the ARSC dataset to study a simple application of transfer learning approaches to few-shot classification. We train a single binary classifier to learn all few-shot classes jointly by prefixing class identifiers to the input text. Given the text and class, the model then makes a binary prediction for that text/class pair. Our results show that this simple approach can outperform most published results on this dataset. Surprisingly, we also show that including domain information as part of the task definition only leads to a modest improvement in model accuracy, and zero-shot classification, without further fine-tuning on few-shot domains, performs equivalently to few-shot classification. These results suggest that the classes in the ARSC few-shot task, which are defined by the intersection of domain and rating, are actually very similar to each other, and that a more suitable dataset is needed for the study of few-shot text classification.
%R 10.18653/v1/2020.coling-main.92
%U https://aclanthology.org/2020.coling-main.92
%U https://doi.org/10.18653/v1/2020.coling-main.92
%P 1061-1066
Markdown (Informal)
[Effective Few-Shot Classification with Transfer Learning](https://aclanthology.org/2020.coling-main.92) (Gupta et al., COLING 2020)
ACL
- Aakriti Gupta, Kapil Thadani, and Neil O’Hare. 2020. Effective Few-Shot Classification with Transfer Learning. In Proceedings of the 28th International Conference on Computational Linguistics, pages 1061–1066, Barcelona, Spain (Online). International Committee on Computational Linguistics.