AutoPSO: A Meta-Framework for Automated Particle Swarm Optimization

2026-07-29

IEEE Transactions on Evolutionary Computation

AI Summary

This work addresses the limitations of traditional particle swarm optimization (PSO) algorithms, which are typically handcrafted, exhibit poor generalization, and lack scalability. To overcome these issues, the authors propose AutoPSO—a meta-framework that automatically constructs customized PSO variants through a bilevel optimization mechanism. In the outer loop, AutoPSO searches over a pool of interchangeable algorithmic components to identify effective combinations; in the inner loop, it leverages EvoX to enable population tensorization and batch evaluation, thereby realizing, for the first time, automated search and flexible replacement within the PSO component space. Experimental results demonstrate that PSO variants discovered by AutoPSO significantly outperform strong baselines on both numerical optimization and neuroevolution-based robot control tasks, with performance consistently improving as population size scales, showcasing superior generalization and extensibility.

Abstract

Particle swarm optimization (PSO) is a widely used metaheuristic, prized for its simplicity and small parameter set. Although decades of research have produced numerous PSO variants that improve performance by modifying key components (e.g., parameter schedules, swarm topologies, or updating rules), two fundamental challenges persist. First, most existing approaches are problem-specific and hand-crafted, leading to poor cross-task generalization and forcing practitioners to navigate an impractically large design space, which also hinders systematic reuse of prior effective mechanisms. Second, mainstream implementations remain CPU-bound, constraining scalability and substantially increasing computational cost in real-world applications. To address these challenges, we propose \emph{AutoPSO}, a highly automated meta-framework for constructing customized PSO algorithms. AutoPSO formulates PSO-based optimization as a bi-level process: an outer search explores the joint space of effective PSO components, while an inner loop instantiates candidate variants to solve the target task and provide feedback. The outer search operates over a curated, open-design component pool, supporting flexible replacement of the component set and the outer optimizer. Crucially, by leveraging EvoX for population tensorization and batched evaluations, AutoPSO can efficiently assess thousands of particles within practical time budgets. Comprehensive experiments on numerical benchmarks and neuroevolution robotic control tasks demonstrate that AutoPSO consistently discovers novel PSO variants that significantly outperform strong baselines. Ablation and scalability studies further highlight the contribution of individual algorithmic components and confirm that AutoPSO achieves increasing performance gains with larger swarm sizes. Source code will be made publicly available.

Problem

• particle swarm optimization
• cross-task generalization
• design space
• computational scalability
• CPU-bound

Innovation

• automated algorithm design
• particle swarm optimization
• bi-level optimization
• scalable evolutionary computation
• component-based metaheuristics