research-article

Designing efficient cascaded classifiers: tradeoff between accuracy and cost

Authors: Vikas C. Raykar, Balaji Krishnapuram, Shipeng YuAuthors Info & Claims

KDD '10: Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining

Pages 853 - 860

Published: 25 July 2010 Publication History

Abstract

We propose a method to train a cascade of classifiers by simultaneously optimizing all its stages. The approach relies on the idea of optimizing soft cascades. In particular, instead of optimizing a deterministic hard cascade, we optimize a stochastic soft cascade where each stage accepts or rejects samples according to a probability distribution induced by the previous stage-specific classifier. The overall system accuracy is maximized while explicitly controlling the expected cost for feature acquisition. Experimental results on three clinically relevant problems show the effectiveness of our proposed approach in achieving the desired tradeoff between accuracy and feature acquisition cost.

Supplementary Material

JPG File (kdd2010_krishnapuram_decc_01.jpg)

Download
13.59 KB

MOV File (kdd2010_krishnapuram_decc_01.mov)

Download
138.66 MB

References

[1]

L. Bourdev and J. Brandt. Robust object detection via soft cascade. In Proceedings of the International Conference on Computer Vision and Pattern Recognition (CVPR), pages 236--243, 2005.

Digital Library

Google Scholar

[2]

C. Capirci, L. Rampin, and et al. Sequential FDG-PET/CT reliably predicts response of locally advanced rectal cancer to neo-adjuvant chemo-radiation therapy. Eur J Nucl Med Mol Imaging, 34:1583--1593, 2007.

Crossref

Google Scholar

[3]

C. Dehing-Oberije, D. De Ruysscher, and et al. Tumor volume combined with number of positive lymph node stations is a more important prognostic factor than TNM stage for survival of non-small-cell lung cancer patients treated with (chemo)radiotherapy. Int J Radiat Oncol Biol Phys., 70(4):1039--1044, 2007.

Crossref

Google Scholar

[4]

M. Dundar and J. Bi. Joint Optimization of Cascaded Classifiers for Computer Aided Detection. In Proceedings of the International Conference on Computer Vision and Pattern Recognition (CVPR), 2007.

Crossref

Google Scholar

[5]

A. Genkin, D. Lewis, and D. Madigan. Large-scale Bayesian logistic regression for text categorization. Technometrics, 49(3):291--304, 2007.

Crossref

Google Scholar

[6]

S. Ji and L. Carin. Cost-sensitive feature acquisition and classification. Pattern Recognition, 40(5):1474--1485, 2007.

Digital Library

Google Scholar

[7]

P. Melville, M. Saar-Tsechansky, F. Provost, and R. Mooney. An expected utility approach to active feature-value acquisition. In ICDM '05: Proceedings of the Fifth IEEE International Conference on Data Mining, pages 745--748, 2005.

Digital Library

Google Scholar

[8]

R. B. Rao, J. Bi, G. Fung, M. Salganicoff, N. Obuchowski, and D. Naidich. LungCAD: a clinically approved, machine learning system for lung cancer detection. In Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 1033--1037, 2007.

Digital Library

Google Scholar

[9]

V. S. Sheng and C. X. Ling. Partial example acquisition in cost-sensitive learning. In KDD '07: Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 638--646, 2007.

Digital Library

Google Scholar

[10]

P. Viola and M. Jones. Rapid object detection using a boosted cascade of simple features. Proceedings of the International Conference on Computer Vision and Pattern Recognition (CVPR), pages 511--518, 2001.

Crossref

Google Scholar

[11]

T. T. Wu and K. Lange. Coordinate descent algorithms for lasso penalized regression. The Annals of Applied Statistics, 2(1):224--244, 2008.

Crossref

Google Scholar

[12]

C. Zhang and P. Viola. Multiple-instance pruning for learning efficient cascade detectors. In J. Platt, D. Koller, Y. Singer, and S. Roweis, editors, Advances in Neural Information Processing Systems 20, pages 1681--1688. MIT Press, Cambridge, MA, 2008.

Google Scholar

[13]

T. Zhang and F. Oles. Text Categorization Based on Regularized Linear Classification Methods. Information Retrieval, 4(1):5--31, 2001.

Digital Library

Google Scholar

Cited By

View all

Xu YKhare AMatlin GRamadoss MKamaleswaran RZhang CTumanov AKoyejo SMohamed SAgarwal ABelgrave DCho KOh A(2022)UnfoldMLProceedings of the 36th International Conference on Neural Information Processing Systems10.5555/3600270.3600602(4598-4611)Online publication date: 28-Nov-2022
https://dl.acm.org/doi/10.5555/3600270.3600602
Birman YHindi SKatz GShabtai A(2022)Cost-effective ensemble models selection using deep reinforcement learningInformation Fusion10.1016/j.inffus.2021.07.01177:C(133-148)Online publication date: 1-Jan-2022
https://dl.acm.org/doi/10.1016/j.inffus.2021.07.011
Ma XWang PZhao HLiu SZhao CLin WLee KXu JZheng BDiaz FShah CSuel TCastells PJones RSakai T(2021)Towards a Better Tradeoff between Effectiveness and Efficiency in Pre-Ranking: A Learnable Feature Selection based ApproachProceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval10.1145/3404835.3462979(2036-2040)Online publication date: 11-Jul-2021
https://dl.acm.org/doi/10.1145/3404835.3462979
Show More Cited By

Index Terms

Designing efficient cascaded classifiers: tradeoff between accuracy and cost
1. Computing methodologies
  1. Machine learning
    1. Learning paradigms
      1. Supervised learning
        Supervised learning by classification
    2. Machine learning approaches
      1. Classification and regression trees

Recommendations

Multi-stage classifier design

In many classification systems, sensing modalities have different acquisition costs. It is often unnecessary to use every modality to classify a majority of examples. We study a multi-stage system in a prediction time cost reduction setting, where the ...
Cost sensitive classification in data mining
ADMA'10: Proceedings of the 6th international conference on Advanced data mining and applications: Part I

Cost-sensitive classification is one of mainstream research topics in data mining and machine learning that induces models from data with unbalance class distributions and impacts by quantifying and tackling the unbalance. Rooted in diagnosis data ...
Energy-efficient amortized inference with cascaded deep classifiers
IJCAI'18: Proceedings of the 27th International Joint Conference on Artificial Intelligence

Deep neural networks have been remarkable successful in various AI tasks but often cast high computation and energy cost for energy-constrained applications such as mobile sensing. We address this problem by proposing a novel framework that optimizes the ...

Comments

Information & Contributors

Information

Published In

KDD '10: Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining

July 2010

1240 pages

ISBN:9781450300551

DOI:10.1145/1835804

General Chairs:
Bharat Rao
Siemens
,
Balaji Krishnapuram
Siemens
,
Program Chairs:
Andrew Tomkins
Google Inc.
,
Qiang Yang
Hong Kong University of Science and Technology

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]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 25 July 2010

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Conference

KDD '10

Sponsor:

KDD '10: The 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

July 25 - 28, 2010

DC, Washington, USA

Acceptance Rates

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

Upcoming Conference

KDD '24

Sponsor:
sigkdd
sigkdd

The 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

August 25 - 29, 2024

Barcelona , Spain

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

22
Total Citations
View Citations
520
Total Downloads

Downloads (Last 12 months)1
Downloads (Last 6 weeks)0

Reflects downloads up to 27 Jul 2024

Other Metrics

View Author Metrics

Citations

Cited By

View all

Xu YKhare AMatlin GRamadoss MKamaleswaran RZhang CTumanov AKoyejo SMohamed SAgarwal ABelgrave DCho KOh A(2022)UnfoldMLProceedings of the 36th International Conference on Neural Information Processing Systems10.5555/3600270.3600602(4598-4611)Online publication date: 28-Nov-2022
https://dl.acm.org/doi/10.5555/3600270.3600602
Birman YHindi SKatz GShabtai A(2022)Cost-effective ensemble models selection using deep reinforcement learningInformation Fusion10.1016/j.inffus.2021.07.01177:C(133-148)Online publication date: 1-Jan-2022
https://dl.acm.org/doi/10.1016/j.inffus.2021.07.011
Ma XWang PZhao HLiu SZhao CLin WLee KXu JZheng BDiaz FShah CSuel TCastells PJones RSakai T(2021)Towards a Better Tradeoff between Effectiveness and Efficiency in Pre-Ranking: A Learnable Feature Selection based ApproachProceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval10.1145/3404835.3462979(2036-2040)Online publication date: 11-Jul-2021
https://dl.acm.org/doi/10.1145/3404835.3462979
Gallagher LChen RBlanco RCulpepper JCulpepper JMoffat ABennett PLerman K(2019)Joint Optimization of Cascade Ranking ModelsProceedings of the Twelfth ACM International Conference on Web Search and Data Mining10.1145/3289600.3290986(15-23)Online publication date: 30-Jan-2019
https://dl.acm.org/doi/10.1145/3289600.3290986
Mahkonen KVirtanen TKämäräinen J(2018)Cascade of Boolean detector combinationsEURASIP Journal on Image and Video Processing10.1186/s13640-018-0303-92018:1Online publication date: 24-Jul-2018
https://doi.org/10.1186/s13640-018-0303-9
Liu SXiao FOu WSi LMatwin SYu SFarooq F(2017)Cascade Ranking for Operational E-commerce SearchProceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining10.1145/3097983.3098011(1557-1565)Online publication date: 13-Aug-2017
https://dl.acm.org/doi/10.1145/3097983.3098011
Chen RGallagher LBlanco RCulpepper JKando NSakai TJoho HLi Hde Vries AWhite R(2017)Efficient Cost-Aware Cascade Ranking in Multi-Stage RetrievalProceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval10.1145/3077136.3080819(445-454)Online publication date: 7-Aug-2017
https://dl.acm.org/doi/10.1145/3077136.3080819
Le LJones D(2017)Guided-Processing Outperforms Duty-Cycling for Energy-Efficient SystemsIEEE Transactions on Circuits and Systems I: Regular Papers10.1109/TCSI.2017.269090964:9(2414-2426)Online publication date: Sep-2017
https://doi.org/10.1109/TCSI.2017.2690909
Le LJones D(2017)Feature-Sharing in Cascade Detection Systems With Multiple ApplicationsIEEE Journal of Selected Topics in Signal Processing10.1109/JSTSP.2017.267953911:3(466-478)Online publication date: Apr-2017
https://doi.org/10.1109/JSTSP.2017.2679539
Robinson DSaria S(2016)Trading-off cost of deployment versus accuracy in learning predictive modelsProceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence10.5555/3060832.3060897(1974-1982)Online publication date: 9-Jul-2016
https://dl.acm.org/doi/10.5555/3060832.3060897
Show More Cited By

View Options

Get Access

Login options

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

Cited By

Index Terms

Recommendations

Multi-stage classifier design

Cost sensitive classification in data mining

Energy-efficient amortized inference with cascaded deep classifiers