Optimizer Benchmarking Needs to Account for Hyperparameter Tuning
This addresses the problem of biased optimizer evaluations for practitioners who face tuning costs, though it is incremental in refining benchmarking practices.
The paper argues that fair optimizer benchmarking must account for the computational cost of hyperparameter tuning, and finds that Adam is the most practical optimizer, especially in low-budget scenarios.
The performance of optimizers, particularly in deep learning, depends considerably on their chosen hyperparameter configuration. The efficacy of optimizers is often studied under near-optimal problem-specific hyperparameters, and finding these settings may be prohibitively costly for practitioners. In this work, we argue that a fair assessment of optimizers' performance must take the computational cost of hyperparameter tuning into account, i.e., how easy it is to find good hyperparameter configurations using an automatic hyperparameter search. Evaluating a variety of optimizers on an extensive set of standard datasets and architectures, our results indicate that Adam is the most practical solution, particularly in low-budget scenarios.