DAMS: Distributed Adaptive Metaheuristic Selection

TL;DR

DAMS is a flexible distributed algorithm for adaptive optimization that coordinates metaheuristics across nodes, improving global performance through local decision-making and adaptive strategies.

Contribution

It introduces a novel distributed adaptive metaheuristic selection framework with a specific implementation called SBM, analyzed both theoretically and experimentally.

Findings

SBM effectively selects the best metaheuristics locally and globally.

DAMS outperforms other adaptive strategies in experiments.

The architecture enables efficient distributed optimization with adaptive decision-making.

Abstract

We present a distributed generic algorithm called DAMS dedicated to adaptive optimization in distributed environments. Given a set of metaheuristic, the goal of DAMS is to coordinate their local execution on distributed nodes in order to optimize the global performance of the distributed system. DAMS is based on three-layer architecture allowing node to decide distributively what local information to communicate, and what metaheuristic to apply while the optimization process is in progress. The adaptive features of DAMS are first addressed in a very general setting. A specific DAMS called SBM is then described and analyzed from both a parallel and an adaptive point of view. SBM is a simple, yet efficient, adaptive distributed algorithm using an exploitation component allowing nodes to select the metaheuristic with the best locally observed performance, and an exploration component allowing nodes to detect the metaheuristic with the actual best performance. The efficiency of BSM-DAMS is demonstrated through experimentations and comparisons with other adaptive strategies (sequential and distributed).