research-article

Improving the Quality of Explanations with Local Embedding Perturbations

Authors:

Kotagiri Ramamohanarao,

Christopher Leckie,

Michael E. HouleAuthors Info & Claims

KDD '19: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining

Pages 875 - 884

https://doi.org/10.1145/3292500.3330930

Published: 25 July 2019 Publication History

Abstract

Classifier explanations have been identified as a crucial component of knowledge discovery. Local explanations evaluate the behavior of a classifier in the vicinity of a given instance. A key step in this approach is to generate synthetic neighbors of the given instance. This neighbor generation process is challenging and it has considerable impact on the quality of explanations. To assess quality of generated neighborhoods, we propose a local intrinsic dimensionality (LID) based locality constraint. Based on this, we then propose a new neighborhood generation method. Our method first fits a local embedding/subspace around a given instance using the LID of the test instance as the target dimensionality, then generates neighbors in the local embedding and projects them back to the original space. Experimental results show that our method generates more realistic neighborhoods and consequently better explanations. It can be used in combination with existing local explanation algorithms.

References

[1]

Philip Adler, Casey Falk, Sorelle A Friedler, Gabriel Rybeck, Carlos Scheidegger, Brandon Smith, and Suresh Venkatasubramanian. 2016. Auditing black-box models for indirect influence. In Proc. 16th IEEE International Conference on Data Mining (ICDM). IEEE, 1--10.

[2]

Laurent Amsaleg, Oussama Chelly, Teddy Furon, Stéphane Girard, Michael E. Houle, Ken-ichi Kawarabayashi, and Michael Nett. 2018. Extreme-value-theoretic estimation of local intrinsic dimensionality. Data Mining and Knowledge Discovery, Vol. 32, 6 (2018), 1768--1805.

Digital Library

[3]

Laurent Amsaleg, Oussama Chelly, Michael E. Houle, Ken-ichi Kawarabayashi, Milovs Radovanović, and Weeris Treeratanajaru. 2019. Intrinsic Dimensionality Estimation within Tight Localities. In Proc. 19th SIAM International Conference on Data Mining (SDM). SIAM, 181--189.

[4]

David Baehrens, Timon Schroeter, Stefan Harmeling, Motoaki Kawanabe, Katja Hansen, and Klaus-Robert Müller. 2010. How to explain individual classification decisions. Journal of Machine Learning Research, Vol. 11, Jun (2010), 1803--1831.

Digital Library

[5]

Nahla Barakat and Joachim Diederich. 2005. Eclectic rule-extraction from support vector machines. Int. Journal of Comp. Intelligence, Vol. 2, 1 (2005), 59--62.

[6]

Mark W. Craven and Jude W. Shavlik. 1996. Extracting tree-structured representations of trained networks. Advances in Neural Information Processing Systems (NIPS) (1996), 24--30.

Digital Library

[7]

Wouter Duivesteijn and Julia Thaele. 2014. Understanding where your classifier does (not) work--the SCaPE model class for EMM. In Proc. 14th IEEE International Conference on Data Mining (ICDM). IEEE, 809--814.

Digital Library

[8]

Ruth C Fong and Andrea Vedaldi. 2017. Interpretable Explanations of Black Boxes by Meaningful Perturbation. In Proc. International Conference on Computer Vision (ICCV). IEEE, 3449--3457.

[9]

Glenn Fung, Sathyakama Sandilya, and R. Bharat Rao. 2005. Rule extraction from linear support vector machines. In Proc. 11th International Conference on Knowledge Discovery and Data Mining (KDD). ACM, 32--40.

Digital Library

[10]

Ian J. Goodfellow, Jonathon Shlens, and Christian Szegedy. 2014. Explaining and Harnessing Adversarial Examples. CoRR (2014).

[11]

Bryce Goodman and Seth Flaxman. 2016. EU regulations on algorithmic decision-making and a “right to explanation”. In ICML Workshop on Human Interpretability in Machine Learning (WHI).

[12]

Andreas Henelius, Kai Puolam"aki, Henrik Boström, Lars Asker, and Panagiotis Papapetrou. 2014. A peek into the black box: exploring classifiers by randomization. Data Mining and Knowledge Discovery, Vol. 28, 5--6 (2014), 1503--1529.

Digital Library

[13]

Andreas Henelius, Kai Puolam"aki, Isak Karlsson, Jing Zhao, Lars Asker, Henrik Boström, and Panagiotis Papapetrou. 2015. Goldeneye

[14]

: A closer look into the black box. In Int. Sym. on Stat. Learning and Data Sciences . Springer, 96--105.

[15]

Michael E. Houle. 2017a. Local intrinsic dimensionality I: an extreme-value-theoretic foundation for similarity applications. In Proc. 10th International Conference on Similarity Search and Applications (SISAP). Springer, 64--79.

[16]

Michael E. Houle. 2017b. Local intrinsic dimensionality II: multivariate analysis and distributional support. In Proc. 10th International Conference on Similarity Search and Applications (SISAP) . Springer, 80--95.

[17]

Franz J. Kurfess. 2000. Neural Networks and Structured Knowledge: Rule Extraction and Applications. Applied Intelligence, Vol. 12, 1 (2000), 7--13.

Digital Library

[18]

Thibault Laugel, Xavier Renard, Marie-Jeanne Lesot, Christophe Marsala, and Marcin Detyniecki. 2018. Defining Locality for Surrogates in Post-hoc Interpretablity. arXiv preprint arXiv:1806.07498 (2018).

[19]

Xingjun Ma, Bo Li, Yisen Wang, Sarah M. Erfani, Sudanthi Wijewickrema, Grant Schoenebeck, Michael E. Houle, Dawn Song, and James Bailey. 2018. Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality. In Proc. 6th International Conference on Learning Representations (ICLR).

[20]

David Martens, Bart Baesens, Tony Van Gestel, and Jan Vanthienen. 2007. Comprehensible credit scoring models using rule extraction from support vector machines. European Journal of Operational Research, Vol. 183, 3 (2007), 1466--1476.

[21]

Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016a. Model-Agnostic Interpretability of Machine Learning. In ICML Workshop on Human Interpretability in Machine Learning (WHI 2016).

[22]

Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016b. Nothing Else Matters: Model-Agnostic Explanations By Identifying Prediction Invariance. In NIPS Workshop on Interpretable Machine Learning in Complex Systems.

[23]

Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016c. Why Should I Trust You?: Explaining the Predictions of Any Classifier. In Proc. 22nd International Conference on Knowledge Discovery and Data Mining (KDD). ACM, 1135--1144.

Digital Library

[24]

Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2018. Anchors: High-precision model-agnostic explanations. In Thirty-Second AAAI Conference on Artificial Intelligence.

[25]

Marko Robnik-vS ikonja and Igor Kononenko. 2008. Explaining classifications for individual instances. IEEE Transactions on Knowledge and Data Engineering, Vol. 20, 5 (2008), 589--600.

Digital Library

[26]

Simone Romano, Oussama Chelly, Xuan Vinh Nguyen, James Bailey, and Michael E. Houle. 2016. Measuring Dependency via Intrinsic Dimensionality. In Proc. 23rd International Conference on Pattern Recognition (ICPR). 1207--1212.

[27]

Alper Kürcs at Uysal. 2018. Feature Selection for Comment Spam Filtering on YouTube. Data Science and Applications, Vol. 1, 1 (2018), 4--8.

[28]

Ziyu Wang, Tom Schaul, Matteo Hessel, Hado Van Hasselt, Marc Lanctot, and Nando De Freitas. 2016. Dueling Network Architectures for Deep Reinforcement Learning. In Proc. 33rd International Conference on Machine Learning (ICML) . JMLR.org, 1995--2003.

Digital Library

Cited By

Yang FZeng GZhong FXiao PZheng WQiu F(2024)CfExplainer: Explainable Just-In-Time Defect Prediction Based on CounterfactualsJournal of Systems and Software10.1016/j.jss.2024.112182(112182)Online publication date: Aug-2024
https://doi.org/10.1016/j.jss.2024.112182
Clifford MErskine JHepburn AFlach PSantos-Rodríguez RFrommholz IHopfgartner FLee MOakes MLalmas MZhang MSantos R(2023)Reconciling Training and Evaluation Objectives in Location Agnostic Surrogate ExplainersProceedings of the 32nd ACM International Conference on Information and Knowledge Management10.1145/3583780.3615284(3833-3837)Online publication date: 21-Oct-2023
https://dl.acm.org/doi/10.1145/3583780.3615284
Nauta MTrienes JPathak SNguyen EPeters MSchmitt YSchlötterer Jvan Keulen MSeifert C(2023)From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AIACM Computing Surveys10.1145/358355855:13s(1-42)Online publication date: 13-Jul-2023
https://dl.acm.org/doi/10.1145/3583558
Show More Cited By

Improving the Quality of Explanations with Local Embedding Perturbations
1. Information systems
  1. Information systems applications

Recommendations

When Should We Use Linear Explanations?
CIKM '22: Proceedings of the 31st ACM International Conference on Information & Knowledge Management

The increasing interest in transparent and fair AI systems has propelled the research in explainable AI (XAI). One of the main research lines in XAI is post-hoc explainability, the task of explaining the logic of an already deployed black-box model. ...
Unraveling the Dilemma of AI Errors: Exploring the Effectiveness of Human and Machine Explanations for Large Language Models
CHI '24: Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems

The field of eXplainable artificial intelligence (XAI) has produced a plethora of methods (e.g., saliency-maps) to gain insight into artificial intelligence (AI) models, and has exploded with the rise of deep learning (DL). However, human-participant ...
Redefining Counterfactual Explanations for Reinforcement Learning: Overview, Challenges and Opportunities
While AI algorithms have shown remarkable success in various fields, their lack of transparency hinders their application to real-life tasks. Although explanations targeted at non-experts are necessary for user trust and human-AI collaboration, the ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image ACM Conferences

KDD '19: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining

July 2019

3305 pages

ISBN:9781450362016

DOI:10.1145/3292500

General Chairs:
Ankur Teredesai
KenSci
,
Vipin Kumar
University of Minnesota
,
Program Chairs:
Ying Li
EV Analysis Corporation
,
Rómer Rosales
LinkedIn
,
Evimaria Terzi
Boston University
,
George Karypis
University of Minnesota

Copyright © 2019 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 25 July 2019

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Funding Sources

JSPS Kakenhi Kiban (B)

Conference

KDD '19

Sponsor:

KDD '19: The 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

August 4 - 8, 2019

AK, Anchorage, USA

Acceptance Rates

KDD '19 Paper Acceptance Rate 110 of 1,200 submissions, 9%;

Overall Acceptance Rate 1,133 of 8,635 submissions, 13%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

17
Total Citations
View Citations
619
Total Downloads

Downloads (Last 12 months)83
Downloads (Last 6 weeks)6

Reflects downloads up to 16 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

Yang FZeng GZhong FXiao PZheng WQiu F(2024)CfExplainer: Explainable Just-In-Time Defect Prediction Based on CounterfactualsJournal of Systems and Software10.1016/j.jss.2024.112182(112182)Online publication date: Aug-2024
https://doi.org/10.1016/j.jss.2024.112182
Clifford MErskine JHepburn AFlach PSantos-Rodríguez RFrommholz IHopfgartner FLee MOakes MLalmas MZhang MSantos R(2023)Reconciling Training and Evaluation Objectives in Location Agnostic Surrogate ExplainersProceedings of the 32nd ACM International Conference on Information and Knowledge Management10.1145/3583780.3615284(3833-3837)Online publication date: 21-Oct-2023
https://dl.acm.org/doi/10.1145/3583780.3615284
Nauta MTrienes JPathak SNguyen EPeters MSchmitt YSchlötterer Jvan Keulen MSeifert C(2023)From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AIACM Computing Surveys10.1145/358355855:13s(1-42)Online publication date: 13-Jul-2023
https://dl.acm.org/doi/10.1145/3583558
Filho RLacerda APappa G(2023)Explainable Regression Via PrototypesACM Transactions on Evolutionary Learning and Optimization10.1145/35769032:4(1-26)Online publication date: 14-Jan-2023
https://dl.acm.org/doi/10.1145/3576903
Alvarez LMenzies T(2023)Don’t Lie to Me: Avoiding Malicious Explanations With STEALTHIEEE Software10.1109/MS.2023.324471340:3(43-53)Online publication date: 26-Apr-2023
https://dl.acm.org/doi/10.1109/MS.2023.3244713
Kuratomi ALee ZMiliou ILindgren TPapapetrou P(2023)ORANGE: Opposite-label soRting for tANGent Explanations in heterogeneous spaces2023 IEEE 10th International Conference on Data Science and Advanced Analytics (DSAA)10.1109/DSAA60987.2023.10302474(1-10)Online publication date: 9-Oct-2023
https://doi.org/10.1109/DSAA60987.2023.10302474
Antoñanzas JJia YFrank EBifet APfahringer B(2023)teex: A toolbox for the evaluation of explanationsNeurocomputing10.1016/j.neucom.2023.126642(126642)Online publication date: Aug-2023
https://doi.org/10.1016/j.neucom.2023.126642
Toliopoulos TGounaris A(2022)Explainable Distance-Based Outlier Detection in Data StreamsIEEE Access10.1109/ACCESS.2022.317234510(47921-47936)Online publication date: 2022
https://doi.org/10.1109/ACCESS.2022.3172345
Ayranci PLai PPhan NHu HKolinowski ANewman DDou D(2022)OnML: an ontology-based approach for interpretable machine learningJournal of Combinatorial Optimization10.1007/s10878-022-00856-z44:1(770-793)Online publication date: 26-Apr-2022
https://doi.org/10.1007/s10878-022-00856-z
Guidotti RMonreale ARuggieri SNaretto FTurini FPedreschi DGiannotti F(2022)Stable and actionable explanations of black-box models through factual and counterfactual rulesData Mining and Knowledge Discovery10.1007/s10618-022-00878-538:5(2825-2862)Online publication date: 14-Nov-2022
https://doi.org/10.1007/s10618-022-00878-5
Show More Cited By

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Media

Figures

Other

Tables

View Table of Contents