Robust Strategic Classification under Decision-Dependent Cost Uncertainty

Sura Alhanouti, Guzin Bayraksan, Parinaz Naghizadeh
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:1880-1908, 2026.

Abstract

Humans facing algorithmic decision systems have been found to “game” them by altering their input data (at a cost to them) in order to favorably change the algorithmic outcomes they receive (at a cost to the algorithm). The growing literature on strategic classification seeks to develop robust machine learning algorithms that account for, and reduce, unwanted strategic behavior. A limitation of these existing works is that they assume the cost of strategic behavior to be fixed and independent of the classifier’s decision. In practice, however, manipulation costs evolve and depend on past algorithmic decisions: today’s decisions influence tomorrow’s costs. This paper proposes and analyzes a two-stage robust optimization framework with a decision-dependent uncertainty set to capture such dependencies. We highlight that awareness of policy-dependent costs not only reduces uncertainty, but also better curtails gaming of the algorithmic system over time.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-alhanouti26a, title = {Robust Strategic Classification under Decision-Dependent Cost Uncertainty}, author = {Alhanouti, Sura and Bayraksan, Guzin and Naghizadeh, Parinaz}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {1880--1908}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/alhanouti26a/alhanouti26a.pdf}, url = {https://proceedings.mlr.press/v306/alhanouti26a.html}, abstract = {Humans facing algorithmic decision systems have been found to “game” them by altering their input data (at a cost to them) in order to favorably change the algorithmic outcomes they receive (at a cost to the algorithm). The growing literature on strategic classification seeks to develop robust machine learning algorithms that account for, and reduce, unwanted strategic behavior. A limitation of these existing works is that they assume the cost of strategic behavior to be fixed and independent of the classifier’s decision. In practice, however, manipulation costs evolve and depend on past algorithmic decisions: today’s decisions influence tomorrow’s costs. This paper proposes and analyzes a two-stage robust optimization framework with a decision-dependent uncertainty set to capture such dependencies. We highlight that awareness of policy-dependent costs not only reduces uncertainty, but also better curtails gaming of the algorithmic system over time.} }
Endnote
%0 Conference Paper %T Robust Strategic Classification under Decision-Dependent Cost Uncertainty %A Sura Alhanouti %A Guzin Bayraksan %A Parinaz Naghizadeh %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-alhanouti26a %I PMLR %P 1880--1908 %U https://proceedings.mlr.press/v306/alhanouti26a.html %V 306 %X Humans facing algorithmic decision systems have been found to “game” them by altering their input data (at a cost to them) in order to favorably change the algorithmic outcomes they receive (at a cost to the algorithm). The growing literature on strategic classification seeks to develop robust machine learning algorithms that account for, and reduce, unwanted strategic behavior. A limitation of these existing works is that they assume the cost of strategic behavior to be fixed and independent of the classifier’s decision. In practice, however, manipulation costs evolve and depend on past algorithmic decisions: today’s decisions influence tomorrow’s costs. This paper proposes and analyzes a two-stage robust optimization framework with a decision-dependent uncertainty set to capture such dependencies. We highlight that awareness of policy-dependent costs not only reduces uncertainty, but also better curtails gaming of the algorithmic system over time.
APA
Alhanouti, S., Bayraksan, G. & Naghizadeh, P.. (2026). Robust Strategic Classification under Decision-Dependent Cost Uncertainty. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:1880-1908 Available from https://proceedings.mlr.press/v306/alhanouti26a.html.

Related Material