A Two-Phase Learning-Based Swarm Optimizer for Large-Scale Optimization.

Area of Science:

• Computational intelligence
• Optimization algorithms
• Swarm intelligence

Background:

• Large-scale optimization problems present significant computational challenges.
• Existing swarm optimization algorithms may struggle with scalability and efficiency.
• Human cooperative learning offers a novel inspiration for algorithm design.

Purpose of the Study:

• To propose a novel swarm optimization algorithm, the Two-Phase Learning-based Swarm Optimizer (TPLSO).
• To enhance the performance of swarm intelligence for large-scale optimization tasks.
• To investigate the exploration and exploitation capabilities of TPLSO.

Main Methods:

• TPLSO incorporates two learning phases: mass learning and elite learning.
• Mass learning involves competitive updates within randomly selected study groups.
• Elite learning facilitates knowledge sharing among high-performing particles.

Main Results:

• Theoretical analysis indicates balanced exploration and exploitation in TPLSO.
• Experimental results on benchmark datasets show TPLSO outperforms several state-of-the-art algorithms.
• TPLSO demonstrates robust performance across diverse large-scale optimization problems.

Conclusions:

• TPLSO is an effective and simple method for large-scale optimization.
• The cooperative learning approach significantly improves swarm optimizer performance.
• TPLSO offers a promising alternative for complex optimization challenges.