@inproceedings{sikdar-etal-2021-integrated,
title = "Integrated Directional Gradients: Feature Interaction Attribution for Neural {NLP} Models",
author = "Sikdar, Sandipan and
Bhattacharya, Parantapa and
Heese, Kieran",
editor = "Zong, Chengqing and
Xia, Fei and
Li, Wenjie and
Navigli, Roberto",
booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.acl-long.71",
doi = "10.18653/v1/2021.acl-long.71",
pages = "865--878",
abstract = "In this paper, we introduce Integrated Directional Gradients (IDG), a method for attributing importance scores to groups of features, indicating their relevance to the output of a neural network model for a given input. The success of Deep Neural Networks has been attributed to their ability to capture higher level feature interactions. Hence, in the last few years capturing the importance of these feature interactions has received increased prominence in ML interpretability literature. In this paper, we formally define the feature group attribution problem and outline a set of axioms that any intuitive feature group attribution method should satisfy. Earlier, cooperative game theory inspired axiomatic methods only borrowed axioms from solution concepts (such as Shapley value) for individual feature attributions and introduced their own extensions to model interactions. In contrast, our formulation is inspired by axioms satisfied by characteristic functions as well as solution concepts in cooperative game theory literature. We believe that characteristic functions are much better suited to model importance of groups compared to just solution concepts. We demonstrate that our proposed method, IDG, satisfies all the axioms. Using IDG we analyze two state-of-the-art text classifiers on three benchmark datasets for sentiment analysis. Our experiments show that IDG is able to effectively capture semantic interactions in linguistic models via negations and conjunctions.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sikdar-etal-2021-integrated">
<titleInfo>
<title>Integrated Directional Gradients: Feature Interaction Attribution for Neural NLP Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sandipan</namePart>
<namePart type="family">Sikdar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Parantapa</namePart>
<namePart type="family">Bhattacharya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kieran</namePart>
<namePart type="family">Heese</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chengqing</namePart>
<namePart type="family">Zong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fei</namePart>
<namePart type="family">Xia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenjie</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roberto</namePart>
<namePart type="family">Navigli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In this paper, we introduce Integrated Directional Gradients (IDG), a method for attributing importance scores to groups of features, indicating their relevance to the output of a neural network model for a given input. The success of Deep Neural Networks has been attributed to their ability to capture higher level feature interactions. Hence, in the last few years capturing the importance of these feature interactions has received increased prominence in ML interpretability literature. In this paper, we formally define the feature group attribution problem and outline a set of axioms that any intuitive feature group attribution method should satisfy. Earlier, cooperative game theory inspired axiomatic methods only borrowed axioms from solution concepts (such as Shapley value) for individual feature attributions and introduced their own extensions to model interactions. In contrast, our formulation is inspired by axioms satisfied by characteristic functions as well as solution concepts in cooperative game theory literature. We believe that characteristic functions are much better suited to model importance of groups compared to just solution concepts. We demonstrate that our proposed method, IDG, satisfies all the axioms. Using IDG we analyze two state-of-the-art text classifiers on three benchmark datasets for sentiment analysis. Our experiments show that IDG is able to effectively capture semantic interactions in linguistic models via negations and conjunctions.</abstract>
<identifier type="citekey">sikdar-etal-2021-integrated</identifier>
<identifier type="doi">10.18653/v1/2021.acl-long.71</identifier>
<location>
<url>https://aclanthology.org/2021.acl-long.71</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>865</start>
<end>878</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Integrated Directional Gradients: Feature Interaction Attribution for Neural NLP Models
%A Sikdar, Sandipan
%A Bhattacharya, Parantapa
%A Heese, Kieran
%Y Zong, Chengqing
%Y Xia, Fei
%Y Li, Wenjie
%Y Navigli, Roberto
%S Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
%D 2021
%8 August
%I Association for Computational Linguistics
%C Online
%F sikdar-etal-2021-integrated
%X In this paper, we introduce Integrated Directional Gradients (IDG), a method for attributing importance scores to groups of features, indicating their relevance to the output of a neural network model for a given input. The success of Deep Neural Networks has been attributed to their ability to capture higher level feature interactions. Hence, in the last few years capturing the importance of these feature interactions has received increased prominence in ML interpretability literature. In this paper, we formally define the feature group attribution problem and outline a set of axioms that any intuitive feature group attribution method should satisfy. Earlier, cooperative game theory inspired axiomatic methods only borrowed axioms from solution concepts (such as Shapley value) for individual feature attributions and introduced their own extensions to model interactions. In contrast, our formulation is inspired by axioms satisfied by characteristic functions as well as solution concepts in cooperative game theory literature. We believe that characteristic functions are much better suited to model importance of groups compared to just solution concepts. We demonstrate that our proposed method, IDG, satisfies all the axioms. Using IDG we analyze two state-of-the-art text classifiers on three benchmark datasets for sentiment analysis. Our experiments show that IDG is able to effectively capture semantic interactions in linguistic models via negations and conjunctions.
%R 10.18653/v1/2021.acl-long.71
%U https://aclanthology.org/2021.acl-long.71
%U https://doi.org/10.18653/v1/2021.acl-long.71
%P 865-878
Markdown (Informal)
[Integrated Directional Gradients: Feature Interaction Attribution for Neural NLP Models](https://aclanthology.org/2021.acl-long.71) (Sikdar et al., ACL-IJCNLP 2021)
ACL