ABSTRACT
We study the design of a class of incentive mechanisms that can effectively prevent cheating in a strategic classification and regression problem. A conventional strategic classification or regression problem is modeled as a Stackelberg game, or a principal-agent problem between the designer of a classifier (the principal) and individuals subject to the classifier's decisions (the agents), potentially from different demographic groups. The former benefits from the accuracy of its decisions, whereas the latter may have an incentive to game the algorithm into making favorable but erroneous decisions. While prior works tend to focus on how to design an algorithm to be more robust to such strategic maneuvering, this study focuses on an alternative, which is to design incentive mechanisms to shape the utilities of the agents and induce effort that genuinely improves their skills, which in turn benefits both parties in the Stackelberg game. Specifically, the principal and the mechanism provider (which could also be the principal itself) move together in the first stage, publishing and committing to a classifier and an incentive mechanism. The agents are (simultaneous) second movers and best respond to the published classifier and incentive mechanism. When an agent's strategic action merely changes its observable features, it hurts the performance of the algorithm. However, if the action leads to improvement in the agent's true label, it not only helps the agent achieve better decision outcomes, but also preserves the performance of the algorithm. We study how a subsidy mechanism can induce improvement actions, positively impact a number of social well-being metrics, such as the overall skill levels of the agents (efficiency) and positive or true positive rate differences between different demographic groups (fairness).
- Mark Braverman and Sumegha Garg. 2020. The Role of Randomness and Noise in Strategic Classification. In 1st Symposium on Foundations of Responsible Computing.Google Scholar
- Michael Brückner, Christian Kanzow, and Tobias Scheffer. 2012. Static Prediction Games for Adversarial Learning Problems. The Journal of Machine Learning Research, Vol. 13 (09 2012), 2617--2654.Google Scholar
- Michael Brückner and Tobias Scheffer. 2011. Stackelberg games for adversarial prediction problems. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 547--555. https://doi.org/10.1145/2020408.2020495Google ScholarDigital Library
- Yatong Chen, Jialu Wang, and Yang Liu. 2020. Strategic Recourse in Linear Classification. arXiv preprint arXiv:2011.00355 (2020).Google Scholar
- Jinshuo Dong, Aaron Roth, Zachary Schutzman, Bo Waggoner, and Zhiwei Steven Wu. 2018. Strategic classification from revealed preferences. In Proceedings of the 2018 ACM Conference on Economics and Computation. 55--70.Google ScholarDigital Library
- Nika Haghtalab, Nicole Immorlica, Brendan Lucier, and Jack Wang. 2020. Maximizing Welfare with Incentive-Aware Evaluation Mechanisms. 160--166. https://doi.org/10.24963/ijcai.2020/23Google Scholar
- Moritz Hardt, Nimrod Megiddo, Christos Papadimitriou, and Mary Wootters. 2016a. Strategic Classification. 111--122. https://doi.org/10.1145/2840728.2840730Google ScholarDigital Library
- Moritz Hardt, Eric Price, and Nati Srebro. 2016b. Equality of opportunity in supervised learning. In Advances in neural information processing systems. 3315--3323.Google Scholar
- Keegan Harris, Hoda Heidari, and Zhiwei Steven Wu. 2021. Stateful Strategic Regression. arxiv: cs.LG/2106.03827Google Scholar
- Lily Hu, Nicole Immorlica, and Jennifer Vaughan. 2019. The Disparate Effects of Strategic Manipulation. 259--268. https://doi.org/10.1145/3287560.3287597Google ScholarDigital Library
- Kun Jin, Tongxin Yin, Charles A. Kamhoua, and Mingyan Liu. 2021. Network Games with Strategic Machine Learning. In Decision and Game Theory for Security, Branislav Bovs anský, Cleotilde Gonzalez, Stefan Rass, and Arunesh Sinha (Eds.). Springer International Publishing, Cham, 118--137.Google Scholar
- Jon Kleinberg and Manish Raghavan. 2020. How Do Classifiers Induce Agents to Invest Effort Strategically? ACM Transactions on Economics and Computation, Vol. 8 (11 2020), 1--23. https://doi.org/10.1145/3417742Google ScholarDigital Library
- John Miller, Smitha Milli, and Moritz Hardt. 2020. Strategic Classification is Causal Modeling in Disguise. In Proceedings of the 37th International Conference on Machine Learning (Proceedings of Machine Learning Research), Hal Daumé III and Aarti Singh (Eds.), Vol. 119. PMLR, 6917--6926.Google Scholar
- Smitha Milli, John Miller, Anca Dragan, and Moritz Hardt. 2019. The Social Cost of Strategic Classification. 230--239. https://doi.org/10.1145/3287560.3287576Google ScholarDigital Library
- US Federal Reserve. 2007. Report to the congress on credit scoring and its effects on the availability and affordability of credit. Board of Governors of the Federal Reserve System (2007).Google Scholar
- Yonadav Shavit, Benjamin Edelman, and Brian Axelrod. 2020. Causal Strategic Linear Regression. arxiv: cs.LG/2002.10066Google Scholar
- Marilyn Strathern. 1997. 'Improving ratings': audit in the British University system. European review, Vol. 5, 3 (1997), 305--321.Google Scholar
Index Terms
- Incentive Mechanisms for Strategic Classification and Regression Problems
Recommendations
The Social Cost of Strategic Classification
FAT* '19: Proceedings of the Conference on Fairness, Accountability, and TransparencyConsequential decision-making typically incentivizes individuals to behave strategically, tailoring their behavior to the specifics of the decision rule. A long line of work has therefore sought to counteract strategic behavior by designing more ...
The Disparate Effects of Strategic Manipulation
FAT* '19: Proceedings of the Conference on Fairness, Accountability, and TransparencyWhen consequential decisions are informed by algorithmic input, individuals may feel compelled to alter their behavior in order to gain a system's approval. Models of agent responsiveness, termed "strategic manipulation," analyze the interaction between ...
Strategic Classification from Revealed Preferences
EC '18: Proceedings of the 2018 ACM Conference on Economics and ComputationWe study an online linear classification problem in which the data is generated by strategic agents who manipulate their features in an effort to change the classification outcome. In rounds, the learner deploys a classifier, then an adversarially ...
Comments