Investigating Grammar-Based Design of Multi-objective Particle Swarm Optimization Algorithm

The importance of automatic algorithms design has been pointed out by researchers. In general, automatically designed algorithms can outperform tailored made ones. This is the case of the Particle Swarm Optimization algorithm (PSO) that has many components that can be chosen such as the velocity equation, etc. Motivated by the success reported by the automatic design of the PSO, this study investigates Multi-objective PSO (MOPSO), an extension of PSO that deals with multi-objective problems. Furthermore, this study presents a framework based on the use of a context-free grammar to guide the design of the MOPSO. The grammar allows the use of different components and parameters from various MOPSOs. Further, the framework offers two design methods: Grammatical Evolution (GE) and Iterated Race (IRACE). Likewise, a set of experiments is made to evaluate the framework using a set of Multi-objective problems, quality indicators and statistical tests. The set of experiments includes the evaluation of: two versions of the grammar, GE against IRACE and a comparison with the Speed-constrained PSO (SMPSO), a well-known multi-objective algorithm.