Reshuffling Resampling Splits Can Improve Generalization of Hyperparameter Optimization
This addresses a practical issue for machine learning practitioners by improving hyperparameter tuning efficiency and performance, though it is incremental as it builds on existing resampling methods.
The paper tackles the problem of hyperparameter optimization by showing that reshuffling resampling splits for each configuration often improves generalization performance on unseen data, with results indicating that reshuffling can make single train-validation holdout competitive with cross-validation while being computationally cheaper.
Hyperparameter optimization is crucial for obtaining peak performance of machine learning models. The standard protocol evaluates various hyperparameter configurations using a resampling estimate of the generalization error to guide optimization and select a final hyperparameter configuration. Without much evidence, paired resampling splits, i.e., either a fixed train-validation split or a fixed cross-validation scheme, are often recommended. We show that, surprisingly, reshuffling the splits for every configuration often improves the final model's generalization performance on unseen data. Our theoretical analysis explains how reshuffling affects the asymptotic behavior of the validation loss surface and provides a bound on the expected regret in the limiting regime. This bound connects the potential benefits of reshuffling to the signal and noise characteristics of the underlying optimization problem. We confirm our theoretical results in a controlled simulation study and demonstrate the practical usefulness of reshuffling in a large-scale, realistic hyperparameter optimization experiment. While reshuffling leads to test performances that are competitive with using fixed splits, it drastically improves results for a single train-validation holdout protocol and can often make holdout become competitive with standard CV while being computationally cheaper.