Projects per year
Abstract
This paper introduces an enhanced metaheuristic (MLACO) that combines machine learning (ML) and ant colony optimization (ACO) to solve combinatorial optimization problems. To illustrate the underlying mechanism of our MLACO algorithm, we start by describing a test problem, the orienteering problem. In this problem, the objective is to find a route that visits a subset of vertices in a graph within a time budget to maximize the collected score. In the first phase of our MLACO algorithm, an ML model is trained using a set of small problem instances where the optimal solution is known. Specifically, classification models are used to classify an edge as being part of the optimal route, or not, using problemspecific features and statistical measures. The trained model is then used to predict the ‘probability’ that an edge in the graph of a test problem instance belongs to the corresponding optimal route. In the second phase, we incorporate the predicted probabilities into the ACO component of our algorithm, i.e., using the probability values as heuristic weights or to warm start the pheromone matrix. Here, the probability values bias sampling towards favoring those predicted ‘highquality’ edges when constructing feasible routes. We have tested multiple classification models including graph neural networks, logistic regression and support vector machines, and the experimental results show that our solution prediction approach consistently boosts the performance of ACO. Further, we empirically show that our ML model trained on small synthetic instances generalizes well to large synthetic and realworld instances. Our approach integrating ML with a metaheuristic is generic and can be applied to a wide range of optimization problems.
Original language  English 

Article number  105769 
Number of pages  16 
Journal  Computers and Operations Research 
Volume  143 
DOIs  
Publication status  Published  Jul 2022 
Keywords
 Ant colony optimization
 Combinatorial optimization
 Machine learning
 Metaheuristic
 Optimal solution prediction
Projects
 1 Finished

Hybrid methods with decomposition for large scale optimization
Li, X., Ernst, A. & Kalyanmoy, D.
16/04/18 → 31/12/20
Project: Research