Abstract
Most optimization problems of practical significance are typically solved by highly configurable parameterized algorithms. To achieve the best performance on a problem instance, a trial-and-error configuration process is required, which is very costly and even prohibitive for problems that are already computationally intensive, e.g. optimization problems associated with machine learning tasks. In the past decades, many studies have been conducted to accelerate the tedious configuration process by learning from a set of training instances. This article refers to these studies as learn to optimize and reviews the progress achieved.
| Original language | English |
|---|---|
| Article number | nwae132 |
| Number of pages | 9 |
| Journal | National Science Review |
| Volume | 11 |
| Issue number | 8 |
| Early online date | 2 Apr 2024 |
| DOIs | |
| Publication status | Published - Aug 2024 |
Bibliographical note
© The Author(s) 2024 Published by Oxford University Press on behalf of China Science Publishing & Media Ltd.Funding
This work was supported by the National Key Research and Development Program of China (2022YFA1004102), the National Natural Science Foundation of China (62250710682) and the Program for Guangdong Introducing Innovative and Entrepreneurial Teams (2017ZT07X386).
Keywords
- optimization
- data-driven algorithm design
- automated algorithm configuration
- machine learning