arXiv:2603.14218v1 Announce Type: new 
Abstract: We study the problem of evaluating the excess risk of large-scale empirical risk minimization under the square loss. Leveraging the idea of wild refitting and resampling, we assume only black-box access to the training algorithm and develop an efficient procedure for estimating the excess risk. Our evaluation algorithm is both computationally and data efficient. In particular, it requires access to only a single dataset and does not rely on any additional validation data. Computationally, it only requires refitting the model on several much smaller datasets obtained through sequential resampling, in contrast to previous wild refitting methods that require full-scale retraining and might therefore be unsuitable for large-scale trained predictors.
  Our algorithm has an interleaved sequential resampling-and-refitting structure. We first construct pseudo-responses through a randomized residual symmetrization procedure. At each round, we thus resample two sub-datasets from the resulting covariate pseudo-response pairs. Finally, we retrain the model separately on these two small artificial datasets. We establish high probability excess risk guarantees under both fixed design and random design settings, showing that with a suitably chosen noise scale, our interleaved resampling and refitting algorithm yields an upper bound on the prediction error. Our theoretical analysis draws on tools from empirical process theory, harmonic analysis, Toeplitz operator theory, and sharp tensor concentration inequalities.

تقدم دراسة جديدة على arXiv طريقة فعالة لتقييم المخاطر الزائدة في تقليل المخاطر التجريبية على نطاق واسع تحت خسارة مربعة، باستخدام تقنيات إعادة العينة وإعادة التهيئة المتداخلة. يسمح هذا النهج بالوصول إلى خوارزميات التدريب كصندوق أسود مع الحاجة فقط إلى مجموعة بيانات واحدة، مما يقلل بشكل كبير من العبء الحسابي مقارنة بالطرق التقليدية.

Un nuevo estudio en arXiv presenta un método eficiente para evaluar el riesgo excesivo de la minimización del riesgo empírico a gran escala bajo pérdida cuadrática, utilizando técnicas de re-muestreo y reajuste entrelazados. Este enfoque permite el acceso de caja negra a los algoritmos de entrenamiento mientras requiere solo un conjunto de datos, reduciendo significativamente la carga computacional en comparación con los métodos tradicionales.

Une nouvelle étude sur arXiv présente une méthode efficace pour évaluer le risque excédentaire de la minimisation du risque empirique à grande échelle sous perte carrée, en utilisant des techniques de rééchantillonnage et de réajustement entrelacés. Cette approche permet un accès en boîte noire aux algorithmes d'entraînement tout en nécessitant seulement un ensemble de données, réduisant ainsi considérablement la charge computationnelle par rapport aux méthodes traditionnelles.

A new study on arXiv presents an efficient method for evaluating the excess risk of large-scale empirical risk minimization under square loss, utilizing interleaved resampling and refitting techniques. This approach allows for black-box access to training algorithms while requiring only a single dataset, significantly reducing the computational burden compared to traditional methods.