A Concentration-Based Artificial Immune Network for Multi-objective Optimization
Until recently, the main focus of researchers that develop algorithms for evolutionary multi-objective optimization has been the creation of mechanisms capable of obtaining sets of solutions that are as close as possible to the true Pareto front of the problem and also as diverse as possible in the objective space, to properly cover such front. However, an adequate maintenance of diversity in the decision space is also important, to efficiently solve several classes of problems and even to facilitate the post-optimization decision making process. This aspect has been widely studied in evolutionary single-objective optimization, what led to the development of several diversity maintenance techniques. Among them, the recently proposed concentration-based artificial immune network (cob-aiNet), which is capable of self-regulating the population size, presented promising results in multimodal problems. So, it is extended here to deal with multi-objective problems that require a proper maintenance of diversity in the decision space.
Keywordsartificial immune systems multi-objective optimization diversity in decision space immune networks
Unable to display preview. Download preview PDF.
- 2.Burnet, F.M.: Clonal selection and after. In: Bell, G.I., Perelson, A.S., Pimgley Jr., G.H. (eds.) Theoretical Immunology, pp. 63–85. Marcel Dekker Inc., New York (1978)Google Scholar
- 4.Chan, K.P., Ray, T.: An evolutionary algorithm to maintain diversity in the parametric and the objective space. In: Proc. of the 2005 Intl. Conference on Computational Intelligence, Robotics and Autonomous Systems, CIRAS 2005 (2005)Google Scholar
- 5.Coelho, G.P., Von Zuben, F.J.: A concentration-based artificial immune network for continuous optimization. In: Proc. of the 2010 IEEE International Conference on Evolutionary Computation, CEC 2010, pp. 108–115 (2010)Google Scholar
- 14.Jerne, N.K.: Towards a network theory of the immune system. Annales d’Immunologie 125(1-2), 373–389 (1974)Google Scholar
- 15.Moore, D.S., McCabe, G.P., Craig, B.: Introduction to the Practice of Statistics, 6th edn. W.H. Freeman, New York (2007)Google Scholar
- 21.Unrich, T., Bader, J., Zitzler, E.: Integrating decision space diversity into hypervolume-based multiobjective search. In: Proc. of the 2010 Genetic and Evolutionary Computation Conference, GECCO 2010 (2010)Google Scholar
- 24.Zitzler, E., Laumanns, M., Thiele, L.: SPEA2: Improving the strength Pareto evolutionary algorithm. Tech. rep., ETH, TIK, Zurich, Switzerland (2001)Google Scholar