Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
80 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
7 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Optimal and Fair Encouragement Policy Evaluation and Learning (2309.07176v3)

Published 12 Sep 2023 in cs.LG and stat.ML

Abstract: In consequential domains, it is often impossible to compel individuals to take treatment, so that optimal policy rules are merely suggestions in the presence of human non-adherence to treatment recommendations. Under heterogeneity, covariates may predict take-up of treatment and final outcome, but differently. While optimal treatment rules optimize causal outcomes across the population, access parity constraints or other fairness considerations on who receives treatment can be important. For example, in social services, a persistent puzzle is the gap in take-up of beneficial services among those who may benefit from them the most. We study causal identification and robust estimation of optimal treatment rules, including under potential violations of positivity. We consider fairness constraints such as demographic parity in treatment take-up, and other constraints, via constrained optimization. Our framework can be extended to handle algorithmic recommendations under an often-reasonable covariate-conditional exclusion restriction, using our robustness checks for lack of positivity in the recommendation. We develop a two-stage algorithm for solving over parametrized policy classes under general constraints to obtain variance-sensitive regret bounds. We illustrate the methods in three case studies based on data from reminders of SNAP benefits recertification, randomized encouragement to enroll in insurance, and from pretrial supervised release with electronic monitoring. While the specific remedy to inequities in algorithmic allocation is context-specific, it requires studying both take-up of decisions and downstream outcomes of them.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (63)
  1. Public safety assessment decision making framework - cook county, il [effective march 2016]. https://news.wttw.com/sites/default/files/article/file-attachments/PSA%20Decision%20Making%20Framework.pdf, 2016.
  2. Dec 2022. URL https://www.whitehouse.gov/wp-content/uploads/2022/12/BurdenReductionStrategies.pdf.
  3. Taming the monster: A fast and simple algorithm for contextual bandits. In International Conference on Machine Learning, pages 1638–1646. PMLR, 2014.
  4. A reductions approach to fair classification. In International Conference on Machine Learning, pages 60–69. PMLR, 2018.
  5. F. Akinnibi and S. Holder. America is the world leader in locking people up. one city found a fix. https://www.bloomberg.com/news/features/2023-08-30/nyc-s-cash-bail-reform-program-is-working-but-caseworkers-need-help, 2023. [Accessed 08-09-2023].
  6. Racial bias in bail decisions. The Quarterly Journal of Economics, 133(4):1885–1932, 2018.
  7. Measuring racial discrimination in bail decisions. American Economic Review, 112(9):2992–3038, 2022.
  8. S. Athey. Beyond prediction: Using big data for policy problems. Science, 2017.
  9. S. Athey and S. Wager. Policy learning with observational data. Econometrica, 89(1):133–161, 2021.
  10. It’s compaslicated: The messy relationship between rai datasets and algorithmic fairness benchmarks. arXiv preprint arXiv:2106.05498, 2021.
  11. Fairness and Machine Learning. fairmlbook.org, 2018. http://www.fairmlbook.org.
  12. P. L. Bartlett and S. Mendelson. Rademacher and gaussian complexities: Risk bounds and structural results. Journal of Machine Learning Research, 3(Nov):463–482, 2002.
  13. Improving human decision-making with machine learning. arXiv preprint arXiv:2108.08454, 2021.
  14. Safe policy learning through extrapolation: Application to pre-trial risk assessment. arXiv preprint arXiv:2109.11679, 2021.
  15. A. Beygelzimer and J. Langford. The offset tree for learning with partial labels. In Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 129–138, 2009.
  16. Improved second-order bounds for prediction with expert advice. Machine Learning, 66:321–352, 2007.
  17. Semi-parametric efficient policy learning with continuous actions. Advances in Neural Information Processing Systems, 32, 2019.
  18. Learning to be fair: A consequentialist approach to equitable decision-making. arXiv preprint arXiv:2109.08792, 2021.
  19. Human capital and administrative burden: The role of cognitive resources in citizen-state interactions. Public Administration Review, 80(1):127–136, 2020.
  20. Counterfactual risk assessments, evaluation, and fairness. In Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency, pages 582–593, 2020.
  21. A case for humans-in-the-loop: Decisions in the presence of erroneous algorithmic scores. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems, pages 1–12, 2020.
  22. Algorithmic risk assessments in the hands of humans. Salem Center, 2020.
  23. The oregon health insurance experiment: evidence from the first year. The Quarterly journal of economics, 127(3):1057–1106, 2012.
  24. D. J. Foster and V. Syrgkanis. Orthogonal statistical learning. arXiv preprint arXiv:1901.09036, 2019.
  25. Y. Freund and R. E. Schapire. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of computer and system sciences, 55(1):119–139, 1997.
  26. B. Green and Y. Chen. Disparate interactions: An algorithm-in-the-loop analysis of fairness in risk assessments. In Proceedings of the conference on fairness, accountability, and transparency, pages 90–99, 2019.
  27. B. Green and Y. Chen. Algorithmic risk assessments can alter human decision-making processes in high-stakes government contexts. Proceedings of the ACM on Human-Computer Interaction, 5(CSCW2):1–33, 2021.
  28. T. Gross. Letter Regarding Electronic Monitoring in Illinois — Community Renewal Society — communityrenewalsociety.org. https://www.communityrenewalsociety.org/blog/letter-regarding-electronic-monitoring-in-illinois. [Accessed 08-09-2023].
  29. P. Herd and D. P. Moynihan. Administrative burden: Policymaking by other means. Russell Sage Foundation, 2019.
  30. Experimental evaluation of algorithm-assisted human decision-making: Application to pretrial public safety assessment. arXiv preprint arXiv:2012.02845, 2020.
  31. Multiply robust estimation of causal effects under principal ignorability. arXiv preprint arXiv:2012.01615, 2020.
  32. N. Kallus and A. Zhou. Confounding-robust policy improvement. In Advances in Neural Information Processing Systems, pages 9269–9279, 2018.
  33. N. Kallus and A. Zhou. Assessing disparate impact of personalized interventions: identifiability and bounds. Advances in neural information processing systems, 32, 2019.
  34. N. Kallus and A. Zhou. Fairness, welfare, and equity in personalized pricing. In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, pages 296–314, 2021a.
  35. N. Kallus and A. Zhou. Minimax-optimal policy learning under unobserved confounding. Management Science, 67(5):2870–2890, 2021b.
  36. Assessing algorithmic fairness with unobserved protected class using data combination. arXiv preprint arXiv:1906.00285, 2019a.
  37. Interval estimation of individual-level causal effects under unobserved confounding. In The 22nd International Conference on Artificial Intelligence and Statistics, pages 2281–2290, 2019b.
  38. Doubly robust counterfactual classification. Advances in Neural Information Processing Systems, 35:34831–34845, 2022.
  39. T. Kitagawa and A. Tetenov. Empirical welfare maximization. 2015.
  40. Does algorithm aversion exist in the field? an empirical analysis of algorithm use determinants in diabetes self-management. An Empirical Analysis of Algorithm Use Determinants in Diabetes Self-Management (July 23, 2021). USC Marshall School of Business Research Paper Sponsored by iORB, No. Forthcoming, 2021.
  41. M. Lipsky. Street-level bureaucracy: Dilemmas of the individual in public service. Russell Sage Foundation, 2010.
  42. Efficient estimation of optimal regimes under a no direct effect assumption. Journal of the American Statistical Association, 116(533):224–239, 2021.
  43. J. Ludwig and S. Mullainathan. Fragile algorithms and fallible decision-makers: lessons from the justice system. Journal of Economic Perspectives, 35(4):71–96, 2021.
  44. The causal impact of bail on case outcomes for indigent defendants in new york city. Observational Studies, 3(1):38–64, 2017.
  45. C. Manski. Social Choice with Partial Knoweldge of Treatment Response. The Econometric Institute Lectures, 2005.
  46. A. Maurer. A vector-contraction inequality for rademacher complexities. In Algorithmic Learning Theory: 27th International Conference, ALT 2016, Bari, Italy, October 19-21, 2016, Proceedings 27, pages 3–17. Springer, 2016.
  47. Offline contextual bandits with high probability fairness guarantees. Advances in neural information processing systems, 32, 2019.
  48. Fairness in risk assessment instruments: Post-processing to achieve counterfactual equalized odds. In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, pages 386–400, 2021.
  49. Office of the Chief Judge. Bail reform in cook county: An examination of general order 18.8a and bail in felony cases. 2019a.
  50. Office of the Chief Judge. Bail reform. 2019b. URL https://www.cookcountycourt.org/HOME/Bail-Reform.
  51. Optimal individualized decision rules using instrumental variable methods. Journal of the American Statistical Association, 116(533):174–191, 2021.
  52. D. B. Rubin. Comments on “randomization analysis of experimental data: The fisher randomization test comment”. Journal of the American Statistical Association, 75(371):591–593, 1980.
  53. Expanding supervised release in new york city. 2022. URL https://safetyandjusticechallenge.org/resources/expanding-supervised-release-in-new-york-city/.
  54. A. Shapiro. On duality theory of conic linear problems. Semi-Infinite Programming: Recent Advances, pages 135–165, 2001.
  55. J. Steinhardt and P. Liang. Adaptivity and optimism: An improved exponentiated gradient algorithm. In International conference on machine learning, pages 1593–1601. PMLR, 2014.
  56. Treatment allocation under uncertain costs. arXiv preprint arXiv:2103.11066, 2021.
  57. A. Swaminathan and T. Joachims. Counterfactual risk minimization. Journal of Machine Learning Research, 2015.
  58. U.S. Commission on Civil Rights. A new paradigm for welfare reform: The need for civil rights enforcement. 2002.
  59. Weak convergence. Springer, 1996.
  60. M. J. Wainwright. High-dimensional statistics: A non-asymptotic viewpoint, volume 48. Cambridge university press, 2019.
  61. Learning non-discriminatory predictors. In Conference on Learning Theory, pages 1920–1953. PMLR, 2017.
  62. “if it didn’t happen, why would i change my decision?”: How judges respond to counterfactual explanations for the public safety assessment. In Proceedings of the AAAI Conference on Human Computation and Crowdsourcing, volume 10, pages 219–230, 2022.
  63. Estimating individualized treatment rules using outcome weighted learning. Journal of the American Statistical Association, 107(499):1106–1118, 2012.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (1)
  1. Angela Zhou (23 papers)
Citations (2)
View on Semantic Scholar