English

Adaptive KDE for Real-Time Thresholding: Prioritized Queues for Financial Crime Investigation

Machine Learning 2026-01-28 v2

Abstract

We study the problem of converting a continuous stream of risk scores into stable decision thresholds under non-stationary score distributions. This problem arises in a wide range of detection systems where scores must be partitioned into prioritized processing regions while preserving semantic consistency over time.

Keywords

Cite

@article{arxiv.2601.14473,
  title  = {Adaptive KDE for Real-Time Thresholding: Prioritized Queues for Financial Crime Investigation},
  author = {Danny Butvinik and Nana Boateng and Achi Hackmon},
  journal= {arXiv preprint arXiv:2601.14473},
  year   = {2026}
}