Clustered Calibration: Representation-Aware Probability Calibration via Learned Subpopulations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lavi, Tomer, Shapira, Bracha, Rappoport, Nadav
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917527396810752
author Lavi, Tomer
Shapira, Bracha
Rappoport, Nadav
author_facet Lavi, Tomer
Shapira, Bracha
Rappoport, Nadav
contents Ensuring that predicted probabilities align with observed frequencies is critical in high-stakes domains such as clinical decision support, autonomous driving and financial risk assessment. Existing calibration methods typically apply a single global transformation or rely on post-hoc binning over predicted confidences, limiting their ability to exploit heterogeneous reliability across sub-populations. We propose Clustered Calibration, a representation-aware framework that identifies sub-populations via clustering in learned feature spaces (e.g., coverage vectors, SHAP values, CNN activations, Transformer embeddings) and fits a soft mixture of cluster-specific parametric calibrators under hierarchical shrinkage toward a global mapping. This design yields context-specific calibration while maintaining global stability. Across six tabular datasets and additional image and text benchmarks, clustered calibration consistently improves or matches strong global calibrators in terms of negative log-likelihood and Brier score, while preserving AUC and accuracy. We further show, both analytically and empirically, that fixed-bin Expected Calibration Error (ECE) can mis-rank soft, region-aware calibrators even when proper scoring rules improve, and we advocate for log-loss and Brier as more reliable bases for model selection in such settings.
format Preprint
id arxiv_https___arxiv_org_abs_2510_19328
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Clustered Calibration: Representation-Aware Probability Calibration via Learned Subpopulations
Lavi, Tomer
Shapira, Bracha
Rappoport, Nadav
Machine Learning
Ensuring that predicted probabilities align with observed frequencies is critical in high-stakes domains such as clinical decision support, autonomous driving and financial risk assessment. Existing calibration methods typically apply a single global transformation or rely on post-hoc binning over predicted confidences, limiting their ability to exploit heterogeneous reliability across sub-populations. We propose Clustered Calibration, a representation-aware framework that identifies sub-populations via clustering in learned feature spaces (e.g., coverage vectors, SHAP values, CNN activations, Transformer embeddings) and fits a soft mixture of cluster-specific parametric calibrators under hierarchical shrinkage toward a global mapping. This design yields context-specific calibration while maintaining global stability. Across six tabular datasets and additional image and text benchmarks, clustered calibration consistently improves or matches strong global calibrators in terms of negative log-likelihood and Brier score, while preserving AUC and accuracy. We further show, both analytically and empirically, that fixed-bin Expected Calibration Error (ECE) can mis-rank soft, region-aware calibrators even when proper scoring rules improve, and we advocate for log-loss and Brier as more reliable bases for model selection in such settings.
title Clustered Calibration: Representation-Aware Probability Calibration via Learned Subpopulations
topic Machine Learning
url https://arxiv.org/abs/2510.19328