The 8088 The 8088 ← All news
arXiv cs.LG AI Research Apr 22

FASE : A Fairness-Aware Spatiotemporal Event Graph Framework for Predictive Policing

★★★★★ significance 3/5

Researchers present FASE, a framework designed to mitigate racial bias in predictive policing by integrating spatiotemporal graph neural networks with fairness-constrained patrol allocation. The study uses Baltimore crime data to demonstrate how spatiotemporal modeling can balance crime risk prediction with demographic impact constraints.

Why it matters Algorithmic interventions in predictive policing represent a critical frontier for mitigating systemic bias in high-stakes automated decision-making.
Read the original at arXiv cs.LG

Tags

#predictive policing #algorithmic fairness #spatiotemporal graphs #bias mitigation

Related coverage