aiqbench analysis
Density-Ratio Rescoring (DRR) presents a novel method for addressing class imbalance by fusing base classifier scores with survey-raking duals, eliminating the need for costly resampling or refitting. Empirical results across 24 tabular benchmarks show consistent improvements in average precision and rare-class ranking, particularly when leveraging random-feature settings. This approach offers a computationally efficient pathway to enhance model performance on imbalanced datasets without altering the underlying training prior.
Density-Ratio Rescoring for Imbalanced Classification
arXiv:2609.23926v1 Announce Type: cross 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.
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.
Machine Learning (stat.ML); Machine Learning (cs.LG)
arXiv:2609.23926 [stat.ML]
(or arXiv:2609.23926v1 [stat.ML] for this version)
https://doi.org/10.48550/arXiv.2609.23926 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 287 words
Related coverage
Robust performance metrics for imbalanced classification problems
arXiv · Aug 25, 2026
Handling Imbalanced Classification: What Works Better Than SMOTE
Analytics Vidhya · Jul 12, 2026
Exploding variance of means of exponentials: least-squares to the rescue – Machine Learning Research Blog
francisbach.com · Sep 27, 2026
DADIR: Density-Aware Data-level Imbalanced Regression Framework
arXiv · Jul 21, 2026