Ant Colony Optimization: A Component-Wise Overview

Abstract

The indirect communication and foraging behavior of certain species of ants have inspired a number of optimization algorithms for NP-hard problems. These algorithms are nowadays collectively known as the ant colony optimization (ACO) metaheuristic. This chapter gives an overview of the history of ACO, explains in detail its algorithmic components, and summarizes its key characteristics. In addition, the chapter introduces a software framework that unifies the implementation of these ACO algorithms for two example problems, the traveling salesman problem and the quadratic assignment problem. By configuring the parameters of the framework, one can combine features from various ACO algorithms in novel ways. Examples on how to find a good configuration automatically are given in the chapter. The chapter closes with a review of combinations of ACO with other techniques and extensions of the ACO metaheuristic to other problem classes.

Solving Sudoku With Ant Colony Optimization

“…In this paper, we focus on the application of ACO to the solution of Sudoku. ACO is a population-based search method inspired by the foraging behaviour of ants [7,9], and it has been successfully applied to a wide range of computational problems (see [6,22] for overviews of both the algorithm and its applications).…”

Myra

“…ACO algorithms have three main caracteristics [12]: (i) they are population-based algorithms, where m ants create solutions to the problem at hand, mimicking a colony of ants; (ii) solutions are created by a probabilistic procedure, where solution components are selected based on pheromone and heuristic information values; (iii) pheromone values are updated at each iteration using the quality of the candidate solutions as (positive) feedback. Figure 1 presents the high-level pseudocode of an ACO algorithm.…”

A distributed algorithm to maintain and repair the trail networks of arboreal ants

“…Ant colony optimization and reinforcement learning. Models of ant colony optimization (ACO), first proposed in 1991, loosely mimic ant behavior to solve combinatorial optimization problems, such as the traveling salesman problem [30,31,32]. In ACO, individual ants each use a heuristic to construct candidate solutions, and then use pheromone to lead other ants towards higher quality solutions.…”