Interventions Over Predictions: Reframing the Ethical Debate for Actuarial Risk Assessment

Proceedings of Machine Learning Research, Forthcoming

9 Pages Posted: 28 Dec 2017

See all articles by Chelsea Barabas

Chelsea Barabas

Massachusetts Institute of Technology (MIT) - MIT Media Laboratory

Karthik Dinakar

Massachusetts Institute of Technology (MIT) - MIT Media Laboratory

Joichi Ito

Massachusetts Institute of Technology (MIT) - MIT Media Laboratory; Harvard Law School

Madars Virza

Massachusetts Institute of Technology (MIT) - MIT Media Laboratory

Jonathan L. Zittrain

Harvard Law School and Harvard Kennedy School of Government; Harvard School of Engineering and Applied Sciences; Berkman Center for Internet & Society; Harvard University - Harvard Kennedy School (HKS)

Date Written: December 21, 2017

Abstract

Actuarial risk assessments might be unduly perceived as a neutral way to counteract implicit bias and increase the fairness of decisions made at almost every juncture of the criminal justice system, from pretrial release to sentencing, parole and probation. In recent times these assessments have come under increased scrutiny, as critics claim that the statistical techniques underlying them might reproduce existing patterns of discrimination and historical biases that are reflected in the data. Much of this debate is centered around competing notions of fairness and predictive accuracy, resting on the contested use of variables that act as “proxies” for characteristics legally protected against discrimination, such as race and gender. We argue that a core ethical debate surrounding the use of regression in risk assessments is not simply one of bias or accuracy. Rather, it’s one of purpose. If machine learning is operationalized merely in the service of predicting individual future crime, then it becomes difficult to break cycles of criminalization that are driven by the iatrogenic effects of the criminal justice system itself. We posit that machine learning should not be used for prediction, but rather to surface covariates that are fed into a causal model for understanding the social, structural and psychological drivers of crime. We propose an alternative application of machine learning and causal inference away from predicting risk scores to risk mitigation.

Keywords: causal inference, criminal justice, interventions, risk assessment tools

Suggested Citation

Barabas, Chelsea and Dinakar, Karthik and Ito, Joichi and Virza, Madars and Zittrain, Jonathan, Interventions Over Predictions: Reframing the Ethical Debate for Actuarial Risk Assessment (December 21, 2017). Proceedings of Machine Learning Research, Forthcoming. Available at SSRN: https://ssrn.com/abstract=3091849

Chelsea Barabas

Massachusetts Institute of Technology (MIT) - MIT Media Laboratory ( email )

20 Ames St.
Cambridge, MA 02139-4307
United States

Karthik Dinakar (Contact Author)

Massachusetts Institute of Technology (MIT) - MIT Media Laboratory ( email )

20 Ames St.
Cambridge, MA 02139-4307
United States

Joichi Ito

Massachusetts Institute of Technology (MIT) - MIT Media Laboratory ( email )

20 Ames St.
Cambridge, MA 02139-4307
United States

Harvard Law School ( email )

1875 Cambridge Street
Cambridge, MA 02138
United States

HOME PAGE: http://https://hls.harvard.edu/faculty/directory/11703/Ito

Madars Virza

Massachusetts Institute of Technology (MIT) - MIT Media Laboratory ( email )

20 Ames St.
Cambridge, MA 02139-4307
United States

Jonathan Zittrain

Harvard Law School and Harvard Kennedy School of Government ( email )

Cambridge, MA 02138
United States

Harvard School of Engineering and Applied Sciences

1875 Cambridge Street
Cambridge, MA 02138
United States

Berkman Center for Internet & Society

Harvard Law School
23 Everett, 2nd Floor
Cambridge, MA 02138
United States

Harvard University - Harvard Kennedy School (HKS) ( email )

79 John F. Kennedy Street
Cambridge, MA 02138
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
168
rank
167,360
Abstract Views
766
PlumX Metrics