Sequential Strategic Screening

被引:0
作者
Cohen, Lee [1 ]
Sharifi-Malvajerdi, Saeed [1 ]
Stangl, Kevin [1 ]
Vakilian, Ali [1 ]
Ziani, Juba [2 ]
机构
[1] Toyota Technol Inst Chicago, Chicago, IL 60637 USA
[2] Georgia Inst Technol, H Milton Stewart Sch Ind & Syst Engn, Atlanta, GA USA
来源
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 202 | 2023年 / 202卷
基金
美国国家科学基金会;
关键词
LEARNING INTERSECTIONS;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We initiate the study of strategic behavior in screening processes with multiple classifiers. We focus on two contrasting settings: a "conjunctive" setting in which an individual must satisfy all classifiers simultaneously, and a sequential setting in which an individual to succeed must satisfy classifiers one at a time. In other words, we introduce the combination of strategic classification with screening processes. We show that sequential screening pipelines exhibit new and surprising behavior where individuals can exploit the sequential ordering of the tests to "zig-zag" between classifiers without having to simultaneously satisfy all of them. We demonstrate an individual can obtain a positive outcome using a limited manipulation budget even when far from the intersection of the positive regions of every classifier. Finally, we consider a learner whose goal is to design a sequential screening process that is robust to such manipulations, and provide a construction for the learner that optimizes a natural objective.
引用
收藏
页数:17
相关论文
共 52 条
  • [1] Ahmadi Saba, 2021, EC '21: Proceedings of the 22nd ACM Conference on Economics and Computation, P6, DOI 10.1145/3465456.3467629
  • [2] Ahmadi S., 2022, S FDN RESP COMP FORC, V218
  • [3] Chasing Convex Bodies with Linear Competitive Ratio
    Argue, C. J.
    Gupta, Anupam
    Tang, Ziye
    Guruganesh, Guru
    [J]. JOURNAL OF THE ACM, 2021, 68 (05)
  • [4] Pipeline Interventions
    Arunachaleswaran, Eshwar Ram
    Kannan, Sampath
    Roth, Aaron
    Ziani, Juba
    [J]. MATHEMATICS OF OPERATIONS RESEARCH, 2022, : 3207 - 3238
  • [5] Bansal N, 2018, SODA'18: PROCEEDINGS OF THE TWENTY-NINTH ANNUAL ACM-SIAM SYMPOSIUM ON DISCRETE ALGORITHMS, P1253
  • [6] Bechavod Y., 2022, INT C MACHINE LEARNI, V162, P1691
  • [7] Bechavod Y, 2021, PR MACH LEARN RES, V130
  • [8] Bjorkegren D., 2020, ARXIV
  • [9] Blum Avrim, 2022, FAccT '22: 2022 ACM Conference on Fairness, Accountability, and Transparency, P1178, DOI 10.1145/3531146.3533178
  • [10] Bower Amanda, 2017, arXiv