Parallelizing Ant Colony Optimization Via Area of Expertise Learning

Ant colony optimization algorithms have long been touted as providing an effective and efficient means of generating high quality solutions to NP-hard optimization problems. Unfortunately, while the structure of the algorithm is easy to parallelize, the nature and amount of communication required for parallel execution has meant that parallel implementations developed suffer from decreased solution quality, slower runtime performance, or both. This thesis explores a new strategy for ant colony parallelization that involves Area of Expertise (AOE) learning. The AOE concept is based on the idea that individual agents tend to gain knowledge of different areas of the search space when left to their own devices. After developing a sense of their own expertness on a portion of the problem domain, agents share information and incorporate knowledge from other agents without having to experience it first-hand. This thesis shows that when incorporated within parallel ACO and applied to multi-objective environments such as a gridworld, the use of AOE learning can be an effective and efficient means of coordinating the efforts of multiple ant colony agents working in tandem, resulting in increased performance. Based on the success of the AOE/ACO combination in gridworld, a similar con guration is applied to the single objective traveling salesman problem. Yet while it was hoped that AOE learning would allow for a fast and beneficial sharing of knowledge between colonies, this goal was not achieved, despite the efforts detailed within. The reason for this lack of performance is due to the nature of the TSP, whose single objective landscape discourages colonies from learning unique portions of the search space. Without this specialization, AOE was found to make parallel ACO faster than the use of a single large colony but less efficient than multiple independent colonies.