Skip to content

A new optimizer using particle swarm theory

Why this mattered

TBD

Abstract

The optimization of nonlinear functions using particle swarm methodology is described. Implementations of two paradigms are discussed and compared, including a recently developed locally oriented paradigm. Benchmark testing of both paradigms is described, and applications, including neural network training and robot task learning, are proposed. Relationships between particle swarm optimization and both artificial life and evolutionary computation are reviewed.

  • citeTHE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMS — Particle swarm optimization cites Fisher's iris measurement paper as a benchmark classification dataset for evaluating the optimizer's search performance.
  • citeParticle swarm optimization — A new optimizer using particle swarm theory builds directly on the original particle swarm optimization algorithm based on social swarm dynamics.
  • enablesParticle swarm optimization — The earlier particle-swarm optimizer supplied the velocity-and-position update mechanism formalized and surveyed in the 2007 particle swarm optimization work.
  • citeParticle swarm optimization — Particle swarm optimization builds on the original particle-swarm optimizer that updates candidate solutions using social and cognitive attraction terms.
  • enablesTHE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMS — Fisher's discriminant analysis introduced multivariate optimization over measured features, a statistical foundation later echoed in population-based search methods like particle swarm optimization.

Sources