Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.05470 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911571629834240 |
|---|---|
| author | Chen, Yuchen Lei, Jing |
| author_facet | Chen, Yuchen Lei, Jing |
| contents | We consider the problem of evaluating black-box multi-class classifiers. In the standard setup, we observe class labels $Y\in \{0,1,\ldots,M-1\}$ generated according to the conditional distribution $
Y|X \sim \text{ Multinom}\big(η(X)\big), $ where $X$ denotes the features and $η$ maps from the feature space to the $(M-1)$-dimensional simplex. A black-box classifier is an estimate $\hatη$ for which we make no assumptions about the training algorithm. Given holdout data, our goal is to evaluate the performance of the classifier $\hatη$. Recent work suggests treating this as a goodness-of-fit problem by testing the hypothesis
$H_0: ρ((X,Y),(X',Y')) \le δ$, where $ρ$ is some metric between two distributions, and $(X',Y')\sim P_X\times \text{ Multinom}(\hatη(X))$. Combining ideas from algorithmic fairness, Neyman-Pearson lemma, and conformal p-values, we propose a new methodology for this testing problem. The key idea is to generate a second sample $(X',Y') \sim P_X \times \text{ Multinom}\big(\hatη(X)\big)$ allowing us to reduce the task to two-sample conditional distribution testing. Using part of the data, we train an auxiliary binary classifier called a distinguisher to attempt to distinguish between the two samples. The distinguisher's ability to differentiate samples, measured using a rank-sum statistic, is then used to assess the difference between $\hatη$ and $η$ . Using techniques from cross-validation central limit theorems, we derive an asymptotically rigorous test under suitable stability conditions of the distinguisher. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_05470 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Evaluating Black-Box Classifiers via Stable Adaptive Two-Sample Inference Chen, Yuchen Lei, Jing Methodology We consider the problem of evaluating black-box multi-class classifiers. In the standard setup, we observe class labels $Y\in \{0,1,\ldots,M-1\}$ generated according to the conditional distribution $ Y|X \sim \text{ Multinom}\big(η(X)\big), $ where $X$ denotes the features and $η$ maps from the feature space to the $(M-1)$-dimensional simplex. A black-box classifier is an estimate $\hatη$ for which we make no assumptions about the training algorithm. Given holdout data, our goal is to evaluate the performance of the classifier $\hatη$. Recent work suggests treating this as a goodness-of-fit problem by testing the hypothesis $H_0: ρ((X,Y),(X',Y')) \le δ$, where $ρ$ is some metric between two distributions, and $(X',Y')\sim P_X\times \text{ Multinom}(\hatη(X))$. Combining ideas from algorithmic fairness, Neyman-Pearson lemma, and conformal p-values, we propose a new methodology for this testing problem. The key idea is to generate a second sample $(X',Y') \sim P_X \times \text{ Multinom}\big(\hatη(X)\big)$ allowing us to reduce the task to two-sample conditional distribution testing. Using part of the data, we train an auxiliary binary classifier called a distinguisher to attempt to distinguish between the two samples. The distinguisher's ability to differentiate samples, measured using a rank-sum statistic, is then used to assess the difference between $\hatη$ and $η$ . Using techniques from cross-validation central limit theorems, we derive an asymptotically rigorous test under suitable stability conditions of the distinguisher. |
| title | Evaluating Black-Box Classifiers via Stable Adaptive Two-Sample Inference |
| topic | Methodology |
| url | https://arxiv.org/abs/2604.05470 |