Improving Fairness in Criminal Justice Algorithmic Risk Assessments Using Conformal Prediction Sets

TL;DR

This paper introduces a conformal prediction framework to improve fairness in criminal justice risk assessments, providing statistically valid, fair forecasts that mitigate racial disparities in offender risk predictions.

Contribution

It presents a novel application of conformal prediction sets to remove unfairness from risk algorithms and covariates, with practical implementation in criminal justice settings.

Findings

Fairness measures show no meaningful differences between Black and White offenders.

The method provides valid probability guarantees for individual risk forecasts.

Procedures are implementable using standard software like R.

Abstract

Risk assessment algorithms have been correctly criticized for potential unfairness, and there is an active cottage industry trying to make repairs. In this paper, we adopt a framework from conformal prediction sets to remove unfairness from risk algorithms themselves and the covariates used for forecasting. From a sample of 300,000 offenders at their arraignments, we construct a confusion table and its derived measures of fairness that are effectively free any meaningful differences between Black and White offenders. We also produce fair forecasts for individual offenders coupled with valid probability guarantees that the forecasted outcome is the true outcome. We see our work as a demonstration of concept for application in a wide variety of criminal justice decisions. The procedures provided can be routinely implemented in jurisdictions with the usual criminal justice datasets used by administrators. The requisite procedures can be found in the scripting software R. However, whether stakeholders will accept our approach as a means to achieve risk assessment fairness is unknown. There also are legal issues that would need to be resolved although we offer a Pareto improvement.