Example-based Explanations with Adversarial Attacks for Respiratory Sound Analysis

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Authors

Research Organisations

External Research Organisations

  • Imperial College London
  • Griffith University Queensland
  • University of Augsburg
View graph of relations

Details

Original languageEnglish
Title of host publicationInterspeech 2022
Subtitle of host publicationIncheon, Korea 18-22 September 2022
Pages4003-4007
Number of pages5
Publication statusPublished - 2022
Event23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022 - Incheon, Korea, Republic of
Duration: 18 Sept 202222 Sept 2022

Publication series

NameProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
ISSN (Print)2308-457X

Abstract

Respiratory sound classification is an important tool for remote screening of respiratory-related diseases such as pneumonia, asthma, and COVID-19. To facilitate the interpretability of classification results, especially ones based on deep learning, many explanation methods have been proposed using prototypes. However, existing explanation techniques often assume that the data is non-biased and the prediction results can be explained by a set of prototypical examples. In this work, we develop a unified example-based explanation method for selecting both representative data (prototypes) and outliers (criticisms). In particular, we propose a novel application of adversarial attacks to generate an explanation spectrum of data instances via an iterative fast gradient sign method. Such unified explanation can avoid over-generalisation and bias by allowing human experts to assess the model mistakes case by case. We performed a wide range of quantitative and qualitative evaluations to show that our approach generates effective and understandable explanation and is robust with many deep learning models.

Keywords

    explainable machine learning, interpretable methods, Respiratory sound analysis

ASJC Scopus subject areas

Sustainable Development Goals

Cite this

Example-based Explanations with Adversarial Attacks for Respiratory Sound Analysis. / Chang, Yi; Ren, Zhao; Nguyen, Thanh Tam et al.
Interspeech 2022 : Incheon, Korea 18-22 September 2022. 2022. p. 4003-4007 (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH).

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Chang, Y, Ren, Z, Nguyen, TT, Nejdl, W & Schuller, BW 2022, Example-based Explanations with Adversarial Attacks for Respiratory Sound Analysis. in Interspeech 2022 : Incheon, Korea 18-22 September 2022. Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, pp. 4003-4007, 23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022, Incheon, Korea, Republic of, 18 Sept 2022. https://doi.org/10.48550/arXiv.2203.16141, https://doi.org/10.21437/Interspeech.2022-11355
Chang, Y., Ren, Z., Nguyen, T. T., Nejdl, W., & Schuller, B. W. (2022). Example-based Explanations with Adversarial Attacks for Respiratory Sound Analysis. In Interspeech 2022 : Incheon, Korea 18-22 September 2022 (pp. 4003-4007). (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH). https://doi.org/10.48550/arXiv.2203.16141, https://doi.org/10.21437/Interspeech.2022-11355
Chang Y, Ren Z, Nguyen TT, Nejdl W, Schuller BW. Example-based Explanations with Adversarial Attacks for Respiratory Sound Analysis. In Interspeech 2022 : Incheon, Korea 18-22 September 2022. 2022. p. 4003-4007. (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH). doi: https://doi.org/10.48550/arXiv.2203.16141, 10.21437/Interspeech.2022-11355
Chang, Yi ; Ren, Zhao ; Nguyen, Thanh Tam et al. / Example-based Explanations with Adversarial Attacks for Respiratory Sound Analysis. Interspeech 2022 : Incheon, Korea 18-22 September 2022. 2022. pp. 4003-4007 (Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH).
Download
@inproceedings{2a2f319f6c5642d8bf14b4c5f5b676b4,
title = "Example-based Explanations with Adversarial Attacks for Respiratory Sound Analysis",
abstract = "Respiratory sound classification is an important tool for remote screening of respiratory-related diseases such as pneumonia, asthma, and COVID-19. To facilitate the interpretability of classification results, especially ones based on deep learning, many explanation methods have been proposed using prototypes. However, existing explanation techniques often assume that the data is non-biased and the prediction results can be explained by a set of prototypical examples. In this work, we develop a unified example-based explanation method for selecting both representative data (prototypes) and outliers (criticisms). In particular, we propose a novel application of adversarial attacks to generate an explanation spectrum of data instances via an iterative fast gradient sign method. Such unified explanation can avoid over-generalisation and bias by allowing human experts to assess the model mistakes case by case. We performed a wide range of quantitative and qualitative evaluations to show that our approach generates effective and understandable explanation and is robust with many deep learning models.",
keywords = "explainable machine learning, interpretable methods, Respiratory sound analysis",
author = "Yi Chang and Zhao Ren and Nguyen, {Thanh Tam} and Wolfgang Nejdl and Schuller, {Bj{\"o}rn W.}",
note = "Funding Information: * Y. Chang and Z. Ren contribute equally. T. T. Nguyen is the corresponding author. This research was partially funded by the Federal Ministry of Education and Research (BMBF), Germany under the project LeibnizKILabor with grant No. 01DD20003 and the research projects “IIP-Ecosphere”, granted by the German Federal Ministry for Economics and Climate Action (BMWK) via funding code No. 01MK20006A. The code is released at https://github.com/ glam-imperial/SoundPrototypeCriticism. ; 23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022 ; Conference date: 18-09-2022 Through 22-09-2022",
year = "2022",
doi = "https://doi.org/10.48550/arXiv.2203.16141",
language = "English",
series = "Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH",
pages = "4003--4007",
booktitle = "Interspeech 2022",

}

Download

TY - GEN

T1 - Example-based Explanations with Adversarial Attacks for Respiratory Sound Analysis

AU - Chang, Yi

AU - Ren, Zhao

AU - Nguyen, Thanh Tam

AU - Nejdl, Wolfgang

AU - Schuller, Björn W.

N1 - Funding Information: * Y. Chang and Z. Ren contribute equally. T. T. Nguyen is the corresponding author. This research was partially funded by the Federal Ministry of Education and Research (BMBF), Germany under the project LeibnizKILabor with grant No. 01DD20003 and the research projects “IIP-Ecosphere”, granted by the German Federal Ministry for Economics and Climate Action (BMWK) via funding code No. 01MK20006A. The code is released at https://github.com/ glam-imperial/SoundPrototypeCriticism.

PY - 2022

Y1 - 2022

N2 - Respiratory sound classification is an important tool for remote screening of respiratory-related diseases such as pneumonia, asthma, and COVID-19. To facilitate the interpretability of classification results, especially ones based on deep learning, many explanation methods have been proposed using prototypes. However, existing explanation techniques often assume that the data is non-biased and the prediction results can be explained by a set of prototypical examples. In this work, we develop a unified example-based explanation method for selecting both representative data (prototypes) and outliers (criticisms). In particular, we propose a novel application of adversarial attacks to generate an explanation spectrum of data instances via an iterative fast gradient sign method. Such unified explanation can avoid over-generalisation and bias by allowing human experts to assess the model mistakes case by case. We performed a wide range of quantitative and qualitative evaluations to show that our approach generates effective and understandable explanation and is robust with many deep learning models.

AB - Respiratory sound classification is an important tool for remote screening of respiratory-related diseases such as pneumonia, asthma, and COVID-19. To facilitate the interpretability of classification results, especially ones based on deep learning, many explanation methods have been proposed using prototypes. However, existing explanation techniques often assume that the data is non-biased and the prediction results can be explained by a set of prototypical examples. In this work, we develop a unified example-based explanation method for selecting both representative data (prototypes) and outliers (criticisms). In particular, we propose a novel application of adversarial attacks to generate an explanation spectrum of data instances via an iterative fast gradient sign method. Such unified explanation can avoid over-generalisation and bias by allowing human experts to assess the model mistakes case by case. We performed a wide range of quantitative and qualitative evaluations to show that our approach generates effective and understandable explanation and is robust with many deep learning models.

KW - explainable machine learning

KW - interpretable methods

KW - Respiratory sound analysis

UR - http://www.scopus.com/inward/record.url?scp=85140094241&partnerID=8YFLogxK

U2 - https://doi.org/10.48550/arXiv.2203.16141

DO - https://doi.org/10.48550/arXiv.2203.16141

M3 - Conference contribution

AN - SCOPUS:85140094241

T3 - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

SP - 4003

EP - 4007

BT - Interspeech 2022

T2 - 23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022

Y2 - 18 September 2022 through 22 September 2022

ER -

By the same author(s)