LOCUS: A Distribution-Free Loss-Quantile Score for Risk-Aware Predictions

TL;DR

Locus is a novel distribution-free method that provides a per-input loss-scale score for risk-aware predictions, enabling effective ranking and control of large-loss events across diverse regression tasks.

Contribution

Introduces Locus, a distribution-free wrapper that models realized loss to produce an interpretable risk score for any fixed prediction function.

Findings

Effective risk ranking across 13 regression benchmarks

Reduces large-loss frequency compared to standard heuristics

Provides distribution-free control of large-loss events

Abstract

Modern machine learning models can be accurate on average yet still make mistakes that dominate deployment cost. We introduce Locus, a distribution-free wrapper that produces a per-input loss-scale reliability score for a fixed prediction function. Rather than quantifying uncertainty about the label, Locus models the realized loss of the prediction function using any engine that outputs a predictive distribution for the loss given an input. A simple split-calibration step turns this function into a distribution-free interpretable score that is comparable across inputs and can be read as an upper loss level. The score is useful on its own for ranking, and it can optionally be thresholded to obtain a transparent flagging rule with distribution-free control of large-loss events. Experiments across 13 regression benchmarks show that Locus yields effective risk ranking and reduces large-loss frequency compared to standard heuristics.