Interleaved Resampling and Refitting: Data and Compute-Efficient Evaluation of Black-Box Predictors

TL;DR

This paper introduces an efficient, data- and compute-friendly method for evaluating the excess risk of large-scale empirical risk minimization models using interleaved resampling and refitting, requiring only a single dataset.

Contribution

It proposes a novel interleaved resampling and refitting algorithm that reduces computational costs and data requirements for excess risk evaluation of black-box predictors.

Findings

The algorithm provides high probability excess risk guarantees.

It requires only small datasets for retraining, avoiding full-scale retraining.

The method is applicable under both fixed and random design settings.

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.