Reproduce “A survey on pioneering metaheuristic algorithms between 2019 and 2024” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Tansel Dokeroglu, Deniz Canturk, Tayfun Kucukyilmaz
This survey identifies and analyzes 23 pioneering metaheuristic algorithms from over 150 new proposals introduced between 2019 and 2024, evaluating them based on criteria such as citation metrics, problem diversity, and innovative mechanisms, and mapping their state-of-the-art applications across various domains.
Key insights
- Many recently proposed metaheuristic algorithms, despite diverse inspirations, often exhibit structural and behavioral similarities to classical algorithms like PSO or GWO.
- A trade-off exists between parameterless algorithms (ease of use) and highly parameterized ones (requiring meticulous tuning for optimal performance).
- Strategies like combining global and local search, introducing randomization, and maintaining population diversity are crucial for escaping local optima.
- Effective binary encoding via transfer functions is vital for applying continuous metaheuristics to discrete and combinatorial optimization problems.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.