research-article

Free access

Strategic classification is causal modeling in disguise

AUTHORs:

Moritz HardtAuthors Info & Claims

ICML'20: Proceedings of the 37th International Conference on Machine Learning

Article No.: 642, Pages 6917 - 6926

Published: 13 July 2020 Publication History

PDF eReader Publisher Site

Abstract

Consequential decision-making incentivizes individuals to strategically adapt their behavior to the specifics of the decision rule. While a long line of work has viewed strategic adaptation as gaming and attempted to mitigate its effects, recent work has instead sought to design classifiers that incentivize individuals to improve a desired quality. Key to both accounts is a cost function that dictates which adaptations are rational to undertake. In this work, we develop a causal framework for strategic adaptation. Our causal perspective clearly distinguishes between gaming and improvement and reveals an important obstacle to incentive design. We prove any procedure for designing classifiers that incentivize improvement must inevitably solve a non-trivial causal inference problem. We show a similar result holds for designing cost functions that satisfy the requirements of previous work. With the benefit of hindsight, our results show much of the prior work on strategic classification is causal modeling in disguise.

Supplementary Material

Additional material (3524938.3525580_supp.pdf)

Supplemental material.

Download
395.08 KB

References

[1]

Alon, T., Dobson, M., Procaccia, A. D., Talgam-Cohen, I., and Tucker-Foltz, J. Multiagent evaluation mechanisms. In AAAI, pp. 1774-1781, 2020.

[2]

Bambauer, J. and Zarsky, T. The algorithm game. Notre Dame L. Rev., 94:1, 2018.

[3]

Bechavod, Y., Ligett, K., Wu, Z. S., and Ziani, J. Causal feature discovery through strategic modification. arXiv preprint arXiv:2002.07024, 2020.

[4]

Belot, M. and Schröder, M. The spillover effects of monitoring: A field experiment. Management Science, 62(1): 37-45, 2016.

[5]

Braverman, M. and Garg, S. The role of randomness and noise in strategic classification. arXiv preprint arXiv:2005.08377, 2020.

[6]

Brückner, M., Kanzow, C., and Scheffer, T. Static prediction games for adversarial learning problems. Journal of Machine Learning Research, 13(Sep):2617-2654, 2012.

[7]

Burrell, J., Kahn, Z., Jonas, A., and Griffin, D. When users control the algorithms: values expressed in practices on twitter. Proceedings of the ACM on Human-Computer Interaction, 3:19, 2019.

Digital Library

[8]

Dalvi, N., Domingos, P., Sanghai, S., Verma, D., et al. Adversarial classification. In Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 99-108. ACM, 2004.

Digital Library

[9]

Dong, J., Roth, A., Schutzman, Z., Waggoner, B., and Wu, Z. S. Strategic classification from revealed preferences. In Proceedings of the 2018 ACM Conference on Economics and Computation, pp. 55-70. ACM, 2018.

Digital Library

[10]

Eberhardt, F., Glymour, C., and Scheines, R. On the number of experiments sufficient and in the worst case necessary to identify all causal relations among n variables. In Proceedings of the Twenty-First Conference on Uncertainty in Artificial Intelligence, pp. 178-184. AUAI Press, 2005.

Digital Library

[11]

Everitt, T., Ortega, P. A., Barnes, E., and Legg, S. Understanding agent incentives using causal influence diagrams, part i: single action settings. arXiv preprint arXiv:1902.09980, 2019.

[12]

Grossman, S. J. and Hart, O. D. An analysis of the principal-agent problem. In Foundations of Insurance Economics, pp. 302-340. Springer, 1992.

[13]

Gubler, T., Larkin, I., and Pierce, L. Motivational spillovers from awards: Crowding out in a multitasking environment. Organization Science, 27(2):286-303, 2016.

Digital Library

[14]

Haghtalab, N., Immorlica, N., Lucier, B., and Wang, J. Maximizing welfare with incentive-aware evaluation mechanisms. In 29th International Joint Conference on Artificial Intelligence, 2020.

[15]

Hand, D., McConway, K., and Stanghellini, E. Graphical models of applicants for credit. IMA Journal of Management Mathematics, 8(2):143-155, 1997.

[16]

Hardt, M., Megiddo, N., Papadimitriou, C., and Wootters, M. Strategic classification. In Proceedings of the 2016 ACM Conference on Innovations in Theoretical Computer Science, pp. 111-122. ACM, 2016.

Digital Library

[17]

Holmstrom, B. and Milgrom, P. Multitask principal-agent analyses: Incentive contracts, asset ownership, and job design. The Journal of Law, Economics, and Organization, 7(special issue):24-52, 1991.

[18]

Hu, L., Immorlica, N., and Vaughan, J. W. The disparate effects of strategic manipulation. In Proceedings of the Conference on Fairness, Accountability, and Transparency, pp. 259-268. ACM, 2019.

Digital Library

[19]

Khajehnejad, M., Tabibian, B., Schölkopf, B., Singla, A., and Gomez-Rodriguez, M. Optimal decision making under strategic behavior. arXiv preprint arXiv:1905.09239, 2019.

[20]

Kleinberg, J. and Raghavan, M. How do classifiers induce agents to invest effort strategically? In Proceedings of the 2019 ACM Conference on Economics and Computation, pp. 825-844. ACM, 2019.

Digital Library

[21]

Milli, S., Miller, J., Dragan, A. D., and Hardt, M. The social cost of strategic classification. In Proceedings of the Conference on Fairness, Accountability, and Transparency, pp. 230-239. ACM, 2019.

Digital Library

[22]

Oates, W. E. and Schwab, R. M. The window tax: A case study in excess burden. Journal of Economic Perspectives, 29(1):163-80, 2015.

[23]

Pearl, J. Causality. Cambridge University Press, 2009.

[24]

Peters, J., Janzing, D., and Schölkopf, B. Elements of Causal Inference: Foundations and Learning Algorithms. Adaptive Computation and Machine Learning series. MIT Press, 2017.

Digital Library

[25]

Rich, J. T. and Larson, J. A. Why some long-term incentives fail. Compensation Review, 16(1):26-37, 1984.

[26]

Ross, S. A. The economic theory of agency: The principal's problem. The American Economic Review, 63(2):134- 139, 1973.

[27]

Shavit, Y., Edelman, B., and Axelrod, B. Learning from strategic agents: Accuracy, improvement, and causality. arXiv preprint arXiv:2002.10066, 2020.

[28]

Shpitser, I. and Pearl, J. Effects of treatment on the treated: identification and generalization. In Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence, pp. 514-521, 2009.

[29]

Ustun, B., Spangher, A., and Liu, Y. Actionable recourse in linear classification. In Proceedings of the Conference on Fairness, Accountability, and Transparency, pp. 10-19. ACM, 2019.

Digital Library

[30]

Ziewitz, M. Rethinking gaming: The ethical work of optimization in web search engines. Social studies of science, pp. 0306312719865607, 2019.

Cited By

Lechner TUrner RBen-David SKrause ABrunskill ECho KEngelhardt BSabato SScarlett J(2023)Strategic classification with unknown user manipulationsProceedings of the 40th International Conference on Machine Learning10.5555/3618408.3619181(18714-18732)Online publication date: 23-Jul-2023
https://dl.acm.org/doi/10.5555/3618408.3619181
Emelianov VGast NLoiseau PPennock DSegal ISeuken S(2022)Fairness in Selection Problems with Strategic CandidatesProceedings of the 23rd ACM Conference on Economics and Computation10.1145/3490486.3538287(375-403)Online publication date: 12-Jul-2022
https://dl.acm.org/doi/10.1145/3490486.3538287
Finocchiaro JMaio RMonachou FPatro GRaghavan MStoica ATsirtsis S(2021)Bridging Machine Learning and Mechanism Design towards Algorithmic FairnessProceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency10.1145/3442188.3445912(489-503)Online publication date: 3-Mar-2021
https://dl.acm.org/doi/10.1145/3442188.3445912
Show More Cited By

Recommendations

Repeated Sales with Multiple Strategic Buyers
EC '17: Proceedings of the 2017 ACM Conference on Economics and Computation

In a market with repeated sales of a single item to a single buyer, prior work has established the existence of a zero revenue perfect Bayesian equilibrium in the absence of a commitment device for the seller. This counter-intuitive outcome is the ...
Computing with strategic agents
Nash Social Welfare Approximation for Strategic Agents
EC '17: Proceedings of the 2017 ACM Conference on Economics and Computation

The fair division of resources among strategic agents is an important age-old problem that has led to a rich body of literature. At the center of this literature lies the question of whether there exist mechanisms that can implement fair outcomes, ...

Comments

Information & Contributors

Information

Published In

cover image Guide Proceedings

ICML'20: Proceedings of the 37th International Conference on Machine Learning

July 2020

11702 pages

Editors:
Hal Daumé,
Aarti Singh

Copyright © 2020.

Publisher

JMLR.org

Publication History

Published: 13 July 2020

Qualifiers

Research-article
Research
Refereed limited

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

4
Total Citations
View Citations
43
Total Downloads

Downloads (Last 12 months)34
Downloads (Last 6 weeks)4

Reflects downloads up to 01 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

Lechner TUrner RBen-David SKrause ABrunskill ECho KEngelhardt BSabato SScarlett J(2023)Strategic classification with unknown user manipulationsProceedings of the 40th International Conference on Machine Learning10.5555/3618408.3619181(18714-18732)Online publication date: 23-Jul-2023
https://dl.acm.org/doi/10.5555/3618408.3619181
Emelianov VGast NLoiseau PPennock DSegal ISeuken S(2022)Fairness in Selection Problems with Strategic CandidatesProceedings of the 23rd ACM Conference on Economics and Computation10.1145/3490486.3538287(375-403)Online publication date: 12-Jul-2022
https://dl.acm.org/doi/10.1145/3490486.3538287
Finocchiaro JMaio RMonachou FPatro GRaghavan MStoica ATsirtsis S(2021)Bridging Machine Learning and Mechanism Design towards Algorithmic FairnessProceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency10.1145/3442188.3445912(489-503)Online publication date: 3-Mar-2021
https://dl.acm.org/doi/10.1145/3442188.3445912
Rosenfeld NHilgard SRavindranath SParkes DLarochelle HRanzato MHadsell RBalcan MLin H(2020)From predictions to decisionsProceedings of the 34th International Conference on Neural Information Processing Systems10.5555/3495724.3496070(4115-4126)Online publication date: 6-Dec-2020
https://dl.acm.org/doi/10.5555/3495724.3496070

View Options

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

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

Media

Figures

Other

Tables

View Table of Contents