Strategic Instrumental Variable Regression: Recovering Causal Relationships From Strategic Responses

被引:0
|
作者
Harris, Keegan [1 ]
Ngo, Daniel [2 ]
Stapleton, Logan [2 ]
Heidari, Hoda [1 ]
Wu, Zhiwei Steven [1 ]
机构
[1] Carnegie Mellon Univ, Sch Comp Sci, Pittsburgh, PA 15213 USA
[2] Univ Minnesota, Comp Sci Dept, Minneapolis, MN USA
基金
美国国家科学基金会;
关键词
IDENTIFICATION;
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
In settings where Machine Learning (ML) algorithms automate or inform consequential decisions about people, individual decision subjects are often incentivized to strategically modify their observable attributes to receive more favorable predictions. As a result, the distribution the assessment rule is trained on may differ from the one it operates on in deployment. While such distribution shifts, in general, can hinder accurate predictions, our work identifies a unique opportunity associated with shifts due to strategic responses: We show that we can use strategic responses effectively to recover causal relationships between the observable features and outcomes we wish to predict, even under the presence of unobserved confounding variables. Specifically, our work establishes a novel connection between strategic responses to ML models and instrumental variable (IV) regression by observing that the sequence of deployed models can be viewed as an instrument that affects agents' observable features but does not directly influence their outcomes. We show that our causal recovery method can be utilized to improve decision-making across several important criteria: individual fairness, agent outcomes, and predictive risk. In particular, we show that if decision subjects differ in their ability to modify non-causal attributes, any decision rule deviating from the causal coefficients can lead to (potentially unbounded) individual-level unfairness.
引用
收藏
页数:21
相关论文
共 50 条
  • [1] Strategic planning financial performance relationships in banks: A causal examination
    Hopkins, WE
    Hopkins, SA
    [J]. STRATEGIC MANAGEMENT JOURNAL, 1997, 18 (08) : 635 - 652
  • [2] Causal Random Forests Model Using Instrumental Variable Quantile Regression
    Chen, Jau-er
    Hsiang, Chen-Wei
    [J]. ECONOMETRICS, 2019, 7 (04)
  • [3] Demystifying Causal Features on Adversarial Examples and Causal Inoculation for Robust Network by Adversarial Instrumental Variable Regression
    Kim, Junho
    Lee, Byung-Kwan
    Ro, Yong Man
    [J]. 2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2023, : 12302 - 12312
  • [4] Linear Regression from Strategic Data Sources
    Gast, Nicolas
    Ioannidis, Stratis
    Loiseau, Patrick
    Roussillon, Benjamin
    [J]. ACM TRANSACTIONS ON ECONOMICS AND COMPUTATION, 2020, 8 (02)
  • [5] Pressures for sustainability and strategic responses on employment relationships: The role of the HR manager
    Bornay-Barrachina, Mar
    Guerrero-Villegas, Jaime
    Lopez-Fernandez, Macarena
    Ruiz-Rodriguez, Margarita
    [J]. EUROPEAN RESEARCH ON MANAGEMENT AND BUSINESS ECONOMICS, 2023, 29 (02)
  • [6] Exploring the blurred nature of strategic linkages across the BSC The relevance of "loose" causal relationships
    Francioli, Francesca
    Cinquini, Lino
    [J]. JOURNAL OF ACCOUNTING AND ORGANIZATIONAL CHANGE, 2014, 10 (04): : 486 - +
  • [7] Microplastic in Dredged Sediments: From Databases to Strategic Responses
    Wilkens, J. L.
    Calomeni-Eck, A. J.
    Boyda, J.
    Kennedy, A.
    McQueen, A. D.
    [J]. BULLETIN OF ENVIRONMENTAL CONTAMINATION AND TOXICOLOGY, 2024, 112 (05)
  • [8] From Instrumental to Expressive The Changing Goals of the US Cold War Strategic Embargo
    Dobson, Alan P.
    [J]. JOURNAL OF COLD WAR STUDIES, 2010, 12 (01) : 98 - 119
  • [9] Variance Estimate of Instrumental Variable Two-Stage Logistic Regression Causal Odds Ratio Estimator
    Cai, Bing
    Ten Have, Thomas R.
    Small, Dylan S.
    [J]. PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, 2012, 21 : 123 - 123
  • [10] Capital Structure as a Strategic Variable: Evidence from Collective Bargaining
    Matsa, David A.
    [J]. JOURNAL OF FINANCE, 2010, 65 (03): : 1197 - 1232