A Particle Filter based Multi-Objective Optimization Algorithm: PFOPS

TL;DR

This paper extends particle filter optimization to handle multi-objective problems by introducing PFOPS, which uses path sampling to balance objectives, validated on diverse benchmark Pareto fronts.

Contribution

It introduces PFOPS, a novel multi-objective particle filter optimization algorithm using path sampling, expanding PFO's applicability to complex multi-objective problems.

Findings

Successfully optimized convex Pareto fronts

Effectively handled concave Pareto fronts

Managed discontinuous Pareto fronts

Abstract

This paper is concerned with a recently developed paradigm for population-based optimization, termed particle filter optimization (PFO). This paradigm is attractive in terms of coherence in theory and easiness in mathematical analysis and interpretation. Current PFO algorithms only work for single-objective optimization cases, while many real-life problems involve multiple objectives to be optimized simultaneously. To this end, we make an effort to extend the scope of application of the PFO paradigm to multi-objective optimization (MOO) cases. An idea called path sampling is adopted within the PFO scheme to balance the different objectives to be optimized. The resulting algorithm is thus termed PFO with Path Sampling (PFOPS). The validity of the presented algorithm is assessed based on three benchmark MOO experiments, in which the shapes of the Pareto fronts are convex, concave and discontinuous, respectively.