Krzysztof Socha

Ant colony optimization for continuous domains

In this paper we present an extension of ant colony optimization (ACO) to continuous domains. We show how ACO, which was initially developed to be a metaheuristic for combinatorial optimization, can be adapted to continuous optimization without any major conceptual change to its structure. We present the general idea, implementation, and results obtained. We compare the results with those reported in the literature for other continuous optimization methods: other ant-related approaches and other metaheuristics initially developed for combinatorial optimization and later adapted to handle the continuous case. We discuss how our extended ACO compares to those algorithms, and we present some analysis of its efficiency and robustness.

A MAX-MIN Ant System for the University Course Timetabling Problem

We consider a simplification of a typical university course timetabling problem involving three types of hard and three types of soft constraints. A MAX-MIN Ant System, which makes use of a separate local search routine, is proposed for tackling this problem. We devise an appropriate construction graph and pheromone matrix representation after considering alternatives. The resulting algorithm is tested over a set of eleven instances from three classes of the problem. The results demonstrate that the ant system is able to construct significantly better timetables than an algorithm that iterates the local search procedure from random starting solutions.

ACO for continuous and mixed-variable optimization

This paper presents how the Ant Colony Optimization (ACO) metaheuristic can be extended to continuous search domains and applied to both continuous and mixed discrete-continuous optimization problems. The paper describes the general underlying idea, enumerates some possible design choices, presents a first implementation, and provides some preliminary results obtained on well-known benchmark problems. The proposed method is compared to other ant, as well as non-ant methods for continuous optimization.

Ant algorithms for the university course timetabling problem with regard to the state-of-the-art

Two ant algorithms solving a simplified version of a typical university course timetabling problem are presented -Ant Colony System and MAX-MIN Ant System. The algorithms are tested over a set of instances from three classes of the problem. Results are compared with recent results obtained with several metaheuristics using the same local search routine (or neighborhood definition), and a reference random restart local search algorithm. Further, both ant algorithms are compared on an additional set of instances. Conclusions are drawn about the performance of ant algorithms on timetabling problems in comparison to other metaheuristics. Also the design, implementation, and parameters of ant algorithms solving the university course timetabling problem are discussed. It is shown that the particular implementation of an ant algorithm has significant influence on the observed algorithm performance.

An ant colony optimization algorithm for continuous optimization: application to feed-forward neural network training

Ant colony optimization (ACO) is an optimization technique that was inspired by the foraging behaviour of real ant colonies. Originally, the method was introduced for the application to discrete optimization problems. Recently we proposed a first ACO variant for continuous optimization. In this work we choose the training of feed-forward neural networks for pattern classification as a test case for this algorithm. In addition, we propose hybrid algorithm variants that incorporate short runs of classical gradient techniques such as backpropagation. For evaluating our algorithms we apply them to classification problems from the medical field, and compare the results to some basic algorithms from the literature. The results show, first, that the best of our algorithms are comparable to gradient-based algorithms for neural network training, and second, that our algorithms compare favorably with a basic genetic algorithm.

An effective hybrid algorithm for university course timetabling

The university course timetabling problem is an optimisation problem in which a set of events has to be scheduled in timeslots and located in suitable rooms. Recently, a set of benchmark instances was introduced and used for an ‘International Timetabling Competition’ to which 24 algorithms were submitted by various research groups active in the field of timetabling. We describe and analyse a hybrid metaheuristic algorithm which was developed under the very same rules and deadlines imposed by the competition and outperformed the official winner. It combines various construction heuristics, tabu search, variable neighbourhood descent and simulated annealing. Due to the complexity of developing hybrid metaheuristics, we strongly relied on an experimental methodology for configuring the algorithms as well as for choosing proper parameter settings. In particular, we used racing procedures that allow an automatic or semi-automatic configuration of algorithms with a good save in time. Our successful example shows that the systematic design of hybrid algorithms through an experimental methodology leads to high performing algorithms for hard combinatorial optimisation problems.

Ant Colony Optimization for Mixed-Variable Optimization Problems

In this paper, we introduce ACOMV: an ant colony optimization (ACO) algorithm that extends the ACOR algorithm for continuous optimization to tackle mixed-variable optimization problems. In ACOMV, the decision variables of an optimization problem can be explicitly declared as continuous, ordinal, or categorical, which allows the algorithm to treat them adequately. ACOMV includes three solution generation mechanisms: a continuous optimization mechanism (ACOR), a continuous relaxation mechanism (ACOMV-o) for ordinal variables, and a categorical optimization mechanism (ACOMV-c) for categorical variables. Together, these mechanisms allow ACOMV to tackle mixed-variable optimization problems. We also define a novel procedure to generate artificial, mixed-variable benchmark functions, and we use it to automatically tune ACOMVs parameters. The tuned ACOMV is tested on various real-world continuous and mixed-variable engineering optimization problems. Comparisons with results from the literature demonstrate the effectiveness and robustness of ACOMV on mixed-variable optimization problems.

Training feed-forward neural networks with ant colony optimization: an application to pattern classification

Ant colony optimization (ACO) is an optimization technique that was inspired by the foraging behaviour of real ant colonies. Originally, the method was introduced for the application to discrete optimization problems. Research efforts led to the development of algorithms for the application to continuous optimization problems. In this paper we extend and apply one of the most successful variants for the training of feed-forward neural networks. For evaluating our algorithm we apply it to pattern classification problems from the medical field. The results show that our algorithm is comparable to specialized algorithms for neural network training, and that it has advantages over other general purpose optimizers.