Density-Ratio Rescoring for Imbalanced Classification Using Raking Duals and Classifier Scores
Abstract
Density-Ratio Rescoring (DRR) augments a classifier trained at the original class prior with a survey-raking dual score. Raking reweights the majority sample to match minority feature moments within a tolerance. DRR marginally standardizes the dual and base scores and combines them with a fixed weight of one half, using the fitted dual directly for prediction without resampling or refitting the base classifier. Under exact population matching and a correctly specified log-linear tilt model, the dual equals the log density ratio up to an additive constant. A class-separation analysis characterizes the signal strength and correlation conditions under which fusion improves separation under common within-class covariance. On 24 tabular benchmarks, evaluated over 30 trials and five base learners, DRR at the D=128 random-feature setting improves average precision over the standardized base on every dataset, with a mean gain of 0.034. It exceeds the shared-dual raking-and-relabeling resampler on 22 of 24 datasets, with a mean gain of $0.092$, and on all eight one-versus-rest tasks of a shared gene-expression cohort. These results demonstrate the effectiveness of using raking duals as reusable scores for improving rare-class ranking while retaining classifiers trained at the original prior.
Related
- Class-Conditioned Diffusion Augmentation and Explainable In-Context Ensemble Learning for Imbalanced Maintenance Action Classification in Aircraft Systems
- Local data augmentation for imbalanced binary classification
- Zero-Complement Extrapolation: Parameter Recovery and Classification in Severely Imbalanced Logistic Regression under Rank-Deficient Regimes
- Hierarchical Majority Decomposition for Imbalanced Classification: A Recursive Binary Majority-Class Tree with Persistent Minority Comparison
- Optimal Two‐Phase Sampling Designs for Generalized Raking Estimators With Multiple Parameters of Interest
BibTeX
@misc{kim2026densityratio,
title = {{Density-Ratio Rescoring for Imbalanced Classification Using Raking Duals and Classifier Scores}},
author = {Dongha Kim and Seunghwan Park},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.23926},
eprint = {2609.23926},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.23926},
}
RIS
TY - UNPB
TI - Density-Ratio Rescoring for Imbalanced Classification Using Raking Duals and Classifier Scores
AU - Dongha Kim
AU - Seunghwan Park
PY - 2026
DA - 2026/09/22
PB - arXiv
DO - 10.48550/arxiv.2609.23926
UR - https://arxiv.org/abs/2609.23926
AB - Density-Ratio Rescoring (DRR) augments a classifier trained at the original class prior with a survey-raking dual score. Raking reweights the majority sample to match minority feature moments within a tolerance. DRR marginally standardizes the dual and base scores and combines them with a fixed weight of one half, using the fitted dual directly for prediction without resampling or refitting the base classifier. Under exact population matching and a correctly specified log-linear tilt model, the dual equals the log density ratio up to an additive constant. A class-separation analysis characterizes the signal strength and correlation conditions under which fusion improves separation under common within-class covariance. On 24 tabular benchmarks, evaluated over 30 trials and five base learners, DRR at the D=128 random-feature setting improves average precision over the standardized base on every dataset, with a mean gain of 0.034. It exceeds the shared-dual raking-and-relabeling resampler on 22 of 24 datasets, with a mean gain of $0.092$, and on all eight one-versus-rest tasks of a shared gene-expression cohort. These results demonstrate the effectiveness of using raking duals as reusable scores for improving rare-class ranking while retaining classifiers trained at the original prior.
ER -