Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 179 - 193
Опубликована: Ноя. 28, 2024
Язык: Английский
Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 179 - 193
Опубликована: Ноя. 28, 2024
Язык: Английский
Опубликована: Янв. 1, 2025
Язык: Английский
Процитировано
0Lecture notes in computer science, Год журнала: 2025, Номер unknown, С. 20 - 35
Опубликована: Янв. 1, 2025
Язык: Английский
Процитировано
0Computer Science Review, Год журнала: 2025, Номер 57, С. 100764 - 100764
Опубликована: Май 22, 2025
Язык: Английский
Процитировано
0Wiley Interdisciplinary Reviews Computational Statistics, Год журнала: 2025, Номер 17(2)
Опубликована: Май 26, 2025
ABSTRACT Benchmarking in optimization is a critical step evaluating the performance, robustness, and scalability of machine learning algorithms metaheuristics. While trends benchmark design continue to evolve, synthetic functions remain vital for fundamental stress tests theoretical evaluations. As several test have been developed derived over past decades, little attention has given classifying such rationale behind their usage. From this lens, paper reviews categorizes broad range often employed assessing optimizers More specifically, we classify based on modality, dimensionality, separability, smoothness, constraints, noise characteristics offer view that aids selecting appropriate benchmarks various algorithmic challenges. Then, review also discusses detail 25 most commonly used open literature proposes two new, highly dimensional, dynamic, challenging could be testing new algorithms. Finally, identifies gaps current benchmarking practices directions future research, as well suggests best guidelines.
Язык: Английский
Процитировано
0Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 179 - 193
Опубликована: Ноя. 28, 2024
Язык: Английский
Процитировано
0