Mind the Gap: Measuring Generalization Performance Across Multiple Objectives
This addresses a methodological gap for researchers and practitioners using MHPO to balance multiple objectives like accuracy and inference time, though it is incremental as it builds on existing MHPO frameworks.
The paper tackles the problem of measuring generalization performance in multi-objective hyperparameter optimization (MHPO) when models may lose Pareto-optimality on the test set, and it introduces a novel evaluation protocol to quantify this performance and compare optimization experiments.
Modern machine learning models are often constructed taking into account multiple objectives, e.g., minimizing inference time while also maximizing accuracy. Multi-objective hyperparameter optimization (MHPO) algorithms return such candidate models, and the approximation of the Pareto front is used to assess their performance. In practice, we also want to measure generalization when moving from the validation to the test set. However, some of the models might no longer be Pareto-optimal which makes it unclear how to quantify the performance of the MHPO method when evaluated on the test set. To resolve this, we provide a novel evaluation protocol that allows measuring the generalization performance of MHPO methods and studying its capabilities for comparing two optimization experiments.